Collaborative Innovation and User Experience Control-Strategies for Monetization of QoS of Data by Cellular

Operators

by

Mohibi Hussain

B.Eng., Electronics/Telecommunications Engineering, University ofEngineering & Technology, Lahore, Pakistan (2001)

Submitted to the Engineering Systems Divisionin partial fulfillment of the requirements for the degree of

Master of Science in Technology and Policy

at the

MASSACHUSETTS INSTITUTE OF TECHNOLOGY

ARCHNESO'F TECHNOLOGy

NOV 2 0 2013

LIBRARIES

September 2013

C Massachusetts Institute of Technology 2013. All rights reserved.

A u th o r...... .................. ..........................................................Engineering Systems Division

Aug 7, 2013

Certified by................. ................ ......... -------------------..-----..................David D. Clark

Senior Research ScientistComputer Science and Artificial Intelligence Laboratory

Thesis Supervisor

Accepted by......Dava J. Newman

Professor of Atronautics and Astronautics and Engineering SystemsDirector, Technology and Policy Program

1

2

Collaborative Innovation and User Experience Control- Strategies forMonetization of QoS of Data by Cellular Operators

byMohibi Hussain

Submitted to the Engineering Systems DivisionOn Aug 7, 2013, in partial fulfillment of the

Requirements for the degree ofMaster of Science in Technology and Policy

Abstract

Monetization of Quality of Service (QoS) for Mobile Data is the prioritization andsegregation of mobile data and charging based on the level of traffic prioritization. Themonetization of QoS of data can be managed by traffic management and performanceoptimization policies. This traffic prioritization can be application specific as well asapplication agnostic, but essentially based on user choice and in compliance with NetNeutrality principles. The mobile operators' quest to evolve from their "dumb bit pipe"role to become more involved in shaping the user experience through collaborativeinnovation with application developers and content providers is the main driver behindthis concept. Three case studies, have been evaluated in this thesis, namely 1)zero.facebook.com, where Facebook collaborates with mobile operators across the worldto bundle a stripped down Facebook application version with a mobile connection, 2)collaboration between Google Voice and Sprint where Sprint is adding Over The TopGoogle Voice minutes to its mobile plans, and 3) NTT DoCoMo's i-mode, which servedas a revolutionary concept in the mobile services, as a traditional mobile operatorreshaped the consumer experience through customized service offerings and by gettinginvolved in almost all sectors of the telecommunication value chain. The commonparameters identified in these case studies, combined with the findings of interviews withover fifty stakeholders across the telecommunications value chain, have been broughttogether in the aggregated model presented in this thesis; "collaborative innovation" and"user experience control" are identified as two significant factors driving the rationalebehind monetization of QoS of mobile data. The regulatory concerns, technologicalaspects, and future scope of this concept are also discussed in this thesis.

Thesis Supervisor: David. D. ClarkTitle: Senior Research ScientistCommunications Future Program

3

Acknowledgements

"Knowledge enlivens the soul"

-Ali Ibn Abu Talib (A.S)

My two years spent at MIT have been a miraculous journey and I am extremely

thankful to my research advisors David Clark and Professor Charles. H. Fine for

providing me with this opportunity of a lifetime. I am grateful to my advisor David Clarkfor including me in the Communications Future Group, and for his invaluable guidance,encouragement and tireless support for pursuing my work. The office space he provided

me in the remarkable Stata Center is undoubtedly my most favorite workplace andoutshines the finest corporate offices around the world with to its unique and inspirational

atmosphere. I would like to thank Professor Charles H. Fine for his brilliant insights,generous support and precious advice throughout my research. He introduced me to theamazing world of Systems Dynamics, which revolutionized my thinking about BusinessSystems. I would also like to thank Natalie Klym, the charismatic executive director ofthe Communications Future Group for her valuable feedback and support.

Since childhood, I was blessed with the unconditional love and guidance of mymaternal grand parents, Saeeda Altaf and Syed Altaf Ali Shah. I am grateful for having a

father who inspired me to become an engineer, my mother who has been the unwaveringpillar of support, my four siblings Zahra, Zaki, Abid and Namvar who keep on makingme proud with their achievements, and my two sons, Murtaza and Moataber, who makemy life worth living every day.

I would like to extend my special gratitude to friends and family who helped me

during the challenging time at MIT, while I was juggling my research and studies with

taking care of my sons. I would specially like to thank my mother Misbah Hussain, my

father S-T- Hussain, my uncle Wasim, my brother Abid, my sister Zahra, and friends like

Shlomiya Bar yam, Dr. Chintan Vaishnav, Susan Perez, Sana Mujahid, Asif Khan, Sabah

Chaudry, Asif Qayyum, Imran Sayeed and Fraz Tajamul for their unconditional supportthrough my journey at MIT.

4

ContentsA b stract .......................................................................................................................................................... 3

Ch apter 1 ..................................................................................................................................................... 10

Introdu ction ............................................................................................................................................... 10

1.1 R esearch M eth odology .................................................................................................................. 12

1.1.1 Interview s ....................................................................................................................................... 12

1.1.2 Case Stu d ies .................................................................................................................................... 12

1.1.3 M od eling .......................................................................................................................................... 13

1.2 T h esis O rgan ization .................................................................................................................... 14

Chapter 2 ..................................................................................................................................................... 15

Zero.faceb ook.com .................................................................................................................................. 15

2.1 Social V iral E ffect Success Story ................................................................................................ is

2.2 Zero.faceb ook.com and M obile O perators ............................................................................ 17

2.3 Causal Loop Diagram for zero.facebook.com ...................................................................... 19

2.4 Lessons Learned from Zero. Facebook. co m .......................................................................... 20

2.4 .1 V alu e for u ser ................................................................................................................................. 20

2.4 .1.a A p p lication A lacrity ....................................................................................................... 20

2.4 .1.b Zero Cost ............................................................................................................................. 21

2.4.2 Value Chain Evolution- Content Delivery Networks ..................................................... 21

2.4 .3 U ser Exp erience Control ........................................................................................................... 22

Ch apter 3 ..................................................................................................................................................... 24

G oogle-V oice an d Sprint ....................................................................................................................... 24

3.1 Lessons learned from Google-Voice and Sprint case study ........................................... 27

3.1.1 C o-op etition .................................................................................................................................... 27

3.1.2 Integration Failures ..................................................................................................................... 27

3.2.3 Negative Feedback from User Experience Control ........................................................ 28

Chapter 4 ..................................................................................................................................................... 29

N T T D oC oM o i-m ode ............................................................................................................................. 29

4.1 NTT-DoCoM o i-m ode Causal Loop Diagram ........................................................................ 34

4 .2 Lesson s Learn ed from i-m ode .................................................................................................... 35

5

4.2.1 Shaping User Experience Control ..................................................................................... 35

4.2.2 Value for User ................................................................................................................................ 35

4.2.3 Collaborative Innovation based on user needs.......................................................... 36

4.2.4 W in-W in arrangement with Content Providers ......................................................... 37

4.2.5 Global Trendsetter.......................................................................................................................38

4.2.6 Point of Failure..............................................................................................................................39

C h a p te r 5 ..................................................................................................................................................... 4 1

The Aggregated Model..........................................................................................................................41

5.1 Parameters Evaluated In The Model .................................................................................. 44

5.2 Intuitions endorsed by the model................................................................................... 48

5.2.1 Collaborative Innovation...................................................................................................... 48

5.2.2 User Experience Control...................................................................................................... 51

C h a p te r 6 ..................................................................................................................................................... 5 4

Rationale Behind Monetization of QoS ..................................................................................... 54

6.1 Mobile Data: Imbalance in demand and capacity ......................................................... 55

6.2 Quality of Service and Quality of User Experience....................................................... 57

6.3 Value Networks and Temporariness of Advantage ...................................................... 59

6.4 Value for the mobile Internet user...................................................................................... 61

C h a p te r 7 ..................................................................................................................................................... 6 5

Regulatory Policy .................................................................................................................................... 65

7.1 Open Internet Rules........................................................................................................................ 66

7.1.a. Transparency .............................................................................................................................. 66

7.1.b. No Blocking................................................................................................................................... 66

7.1.c. No Unreasonable Discrimination .................................................................................. 66

7.2 Regulatory Frameworks for Monetization of QoS....................................................... 72

7.2.1 Theoretical/Open Market Regulatory Framework ................................................... 73

7.2.2 Antitrust Regulatory Framework...................................................................................... 74

C h a p te r 8 ..................................................................................................................................................... 7 6

Technology Aspects of Monetization of QoS .......................................................................... 76

8.1 Internet QoS Today ......................................................................................................................... 78

8.2 Resource Assurance & Performance Optimization....................................................... 79

8.3 Mobile Internet QoS........................................................................................................................79

8.3.1 Technology Evolution of Mobile Standards.................................................................. 79

8.3.2. Long Term Evolution (LTE)....................................................................................................82

6

8.3.3 Cognitive Radio Adm ission Control Schem e ............................................................... 86

8.3.4 N etw ork and Device aw are Quality of Service .......................................................... 86

8.3.5 Interactive M obile M ultim edia Services........................................................................ 86

8.4 Bearers & QoS - EPS QoS Param eters................................................................................ 88

8.4.1 Default & Dedicated Bearer ................................................................................................. 88

8.4.2 GBR Vs. N on-GBR ...................................................................................................................... 88

8.4.3 QCI - QoS Class Identifier ................................................................................................... 88

8.4.4 PCRF (Policy & Charging Rules Functions) .................................................................. 89

8.5 N etw ork Application Plug-Ins (A PIs)................................................................................ 89

Chapter 9.....................................................................................................................................................91

Future W ork .............................................................................................................................................. 91

9.1 Collaborative Innovation & User Experience Control.................................................. 91

9.2 LTE & Sm all Cell M obile Technology................................................................................... 93

9.3 QoE M anagem ent Fram ew ork .............................................................................................. 94

A PPEN DIX A ............................................................................................................................................... 97

A PPEN DIX B............................................................................................................................................100

A PPEN DIX C ............................................................................................................................................ 105

References ............................................................................................................................................... 111

7

List of Figures

2-1 Causal loop diagram for Zero.facebook.com...................................... 19

3-1 Google-Voice & Sprint advert on Google.com & Sprint.com................... 25

3-2 Causal loop diagram for Google-Voice and Sprint............................... 26

4-1 i-m ode business m odel................................................................ 31

4-2 Evolution of mobile web standards.................................................. 32

4-3 NTT DoCoMo i-mode causal loop diagram....................................... 34

5-1 The Aggregated Model with loops.................................................. 43

5-2 The Aggregated Model with Shadow variables to improve visibility of

param eter dependence................................................................ 44

5-3 Scenario 1- Barriers to new entrants- via consumer stickiness.................. 50

5-4 Scenario 2- Barriers to new entrants- via consumer stickiness.................. 51

5-5 Graphical representation of Platform Enrichment, Perception of Value for

Money, User Experience Control, Perception of Intrusiveness over a period

of 20 years.............................................................................. 52

6-1 Perceptual QoS factors affecting the perceived quality of advanced mobile

and IP-based communication......................................................... 58

6-2 The emerging value network of the Mobile industry.............................. 60

6-3 Location based usage of mobile Internet and mobile video...................... 62

8-1 National Research Council, Realizing the Information Future.................. 77

8-2 Evolution from GSM to LTE......................................................... 83

8

List of Tables

Table 8-1: Applications and corresponding QoS requirements...........................78

Table 8-2: The Generations of 3GPP Systems..................................................80

Table 8-3: 3GPP Release-99 to Release-10..................................................81

Table 8-4: Bearers & QoS: QCI Service Characteristics.......................................89

Table 8-5: Examples of APIs Operator/Developer Programs..................................90

9

Chapter 1

Introduction

"To raise new questions, new possibilities, to regard old problems from a new angle,

requires creative imagination and marks real advance in science."

- Albert Einstein

As Albert Einstein rightfully stated about the "real advance in science", I believe

that the same notion applies to the strategy of business (which may be termed as the

science of business management) and to the conception of new technology leading to or

following the business strategy.

While working for and helping major mobile operators strategize their future for

over a decade, I repeatedly encountered the same critical question; how can mobile

operators maximize their profits by experimenting with innovative value-added services

in order to maintain sustainable growth? Upon my arrival at MIT, I started evaluating the

telecommunications value chain and started interviewing stakeholders from the

telecommunications industry; the same question kept surfacing in almost every

discussion.

The marvels of telecommunications industry are not just limited to the fascinating

technological advancements but are also shaping the human social fabric, politics,

business, and the national and global development. The telecommunications industry has

been the most dynamic and volatile during the last three decades. Evolution of mobile

technology has transformed the pattern of human communication across the world. The

access of Internet through mobile phones has raised the usage and reach of Internet to a

colossal scale.

10

The wireless ecosystem is more complex than the wired Internet. All of the key

players in the ecosystem such as service providers or bit-pipe providers, end-user device

manufacturers, back end equipment providers, content providers and application

developers, as well as the end users are all interdependent. This diverse ecosystem

encourages new technological and business joint ventures; collaborative innovations to

enable new service offerings resulting in platform enrichment, and in turn more value to

the consumers. This complexity presents new challenges, effects, and remunerations. It

also demands revision of business models, identification of collaborative innovation

opportunities, and strategic evolution of the value chain players.

This thesis has been conceived and presented in layers. The background layer is

the analysis and study of the existing and evolving value chain of the telecommunications

industry which includes device manufacturers, content providers, content aggregators,

fixed line and wireless communication providers as well as the Internet layer. In the

foundation layer, the thesis uses three case studies to extract lessons into collaborative

innovation and user experience control. The analysis of the profitability of service

providers, driven by collaborative and firm innovation and affected by perception of

value by the user and perception of intrusiveness is deciphered by an aggregated model in

order to look closely into the effects of these drivers on any specific collaborative

innovation in general and the monetization of QoS of data by cellular operators in

particular. The second layer is the analysis of a very narrowly focused idea; monetization

of Quality of Service (QoS) of mobile Internet by cellular operators, and explores the

regulatory barriers and technological realities and forecasts associated with this idea.

It is essential to note that the inferences from this thesis are takeaways for any

player in the value chain of the telecommunications industry. The monetization of QoS of

data by cellular operators is described in particular as one of the possible applications of

these lessons.

11

This thesis explores the following four salient aspects associated with the

monetization of QoS:

1) The Value Chain Evolution via collaborative Innovation & co-opetition

2) User Experience Control

3) The Regulatory Barriers

4) The Technology

1.1 Research Methodology

1.1.1 InterviewsAround 50 formal and informal interviews were conducted with various players and

stakeholders across the telecommunications value chain from US & Europe, The

interviewees included Chief Operating Officers, Strategy Heads, managers, research lab

personnel, operational engineers and application developers from service providers,

wireless and wireline telecom operators, content developers, infrastructure providers and

device manufacturers. The questions were aimed to gauge the direction and trends of the

telecommunications industry in general and the experimentation and innovation in the

technology and services areas of the mobile operators in particular. Detailed

questionnaires used in the interviews can be found in the Appendix section of the thesis.

1.1.2 Case Studies

Three case studies were used to analyze the common parameters and in particular, the

"User Experience Control" and "Collaborative Innovation" as common themes.

12

Though individually unique and mutually disparate, the following three case studies

forming the premise of this research work have been chosen in an attempt to represent the

entire spectrum of the diverse issues pertinent to the question and rationale behind

monetization of Quality of Service by cellular operators based on the ideas of "User

Experience Control" and "Collaborative Innovation".

i. Zero.facebook.com

ii. NTT DoCoMo's i-mode

iii. Sprint-Google Voice partnership

Google voice Sprint.X

1.1.3 Modeling

This thesis employs System Dynamics Modeling, both for individual cases as well as the

aggregated model. The causal loop diagrams that have been simplified and trimmed to

capture significant aspects of the three individual cases studies are presented in Chapters

2,3 and 4. The aggregated model as presented in Chapter 5 has been used to test the

conclusions and endorse / dismiss the intuitions governed by the common parameters

extracted from the three case studies, and to make inferences and identify policy levers.

13

1.2 Thesis Organization

This thesis has been organized in nine chapters. The first chapter is oriented around the

introduction of the question being raised and address by this thesis. Three case studies are

evaluated in this thesis in Chapters 2, 3 and 4. The second chapter is focused on the

Zero.facebook.com case study and presents the causal loop diagram based on the case

study. The third chapter discusses the Google Voice-Sprint case study and the causal loop

diagram, which identifies the parameters salient to the case study for further utilization in

the aggregated model. The fourth chapter analyzes the NTT DoCoMo i-mode case study

and its represented causal loop diagram. The fifth chapter presents the aggregated model

combining the overlapping and unique parameters identified in the three cases studies in

prior chapters. This chapter also describes the parameters of "Collaborative Innovation"

and "User Experience Control" further. The sixth chapter connects the premises of

collaborative innovation and user experience control to justify the rationale behind

monetization of the Quality of Service by cellular / mobile operators. The sixth chapter

also briefly touches upon the difference between the concepts of Quality of Service as

discussed in this thesis, and the Quality of Experience, which is further utilized in the last

chapter. The seventh chapter is based on the regulatory debates around the monetization

of QoS of Internet in general and mobile data in particular. The eighth chapter outlines

the technological aspects of the monetization of QoS of mobile data with some focus on

Long Term Evolution (LTE) of mobile technology. The final chapter sums up some of

the inferences from the work presented and points towards the possible future extensions

from this research. First of the two Appendices in the end contains the questionnaire

utilized for the research mechanism while addressing the question presented by this

thesis. The second Appendix shows causal tree connections drawn from the aggregated

systems dynamics model presented in the fifth chapter.

14

Chapter 2

Zero.facebook.com

"Facebook is bringing the world together.

- David Kirkpatrick, 2010.

Hannu Verkasalo, SVP of the Arbitron Mobile Trends app, stated in Feb 2012

that "the growth of Facebook cannot only be explained by organic growth, but by the

various partnerships and deals Facebook has with carriers and device manufacturers,

which guarantees them with top position in Android and iPhone devices".

The zero.facebook.com initiative is Facebook's attempt at trying to replicate the

initial viral effect success story of Facebook.

2.1 Social Viral Effect Success Story

The original reason for Facebook's quick success at Harvard was that it provided

the students a platform to interact and connect, filling the void of being away from their

families and surrounded by new people at school from diverse backgrounds. The unique

attributes of this revolutionary social networking website were that one could see a

stranger's name and picture, personal information such as the courses they were taking,

their place of origin and above all, about their interests without getting into the social

requirements of physically being present in a night club, bar, speed dating or school

dating forum, or an activity club. A substantial amount of information could be acquired

about a potential friend in a fast, hassle-free and convenient way through the Internet.

(Kirkpatrick, 2010)

15

Facebook's inception at Harvard provided it the prestige associated with the

historic institution. Harvard connection proved to be instrumental in gaining the attention

of students at other prestigious institutions, some of which were already making use of

indigenous online social networking tools, although not quite at par with Facebook.

College students were now demanding Facebook, and it became a hit at other prestigious

colleges too. The phenomenon started spreading to other colleges as well and demand for

Facebook continued to grow. Mark Zuckerburg, founder of Facebook, was now faced

with the question of allowing Facebook access to students outside the Ivy League

institutions; the first critical issue faced by the novice enterprise as its association with

elite colleges served as a brand recognition tool. Going against the concern of its

associates regarding the potential devaluation of the Facebook brand, Zuckerberg decided

to open Facebook to all college students, and subsequently to high school students. His

approach worked as Facebook took off. (Kirkpatrick, 2010)

Soon after, Facebook management faced the decision of ending the exclusivity of

the social network by providing access to everyone; a decision that could make or break

the new enterprise due to issues of brand dilution, server space and soaring costs

associated with the expansion. Existing advertising revenue was not sufficient enough to

cover the substantial costs of expanding the network, so Facebook allowed access to

anyone over thirteen years of age who wanted to join the hip social network. Soon after,

anyone was allowed to write programs that would run on Facebook. There was no fee or

permission required to write and run these applications on Facebook. Thousands of

applications thus developed kept pace with the expanding network. This decision and

value-addition added an extra dimension to ways in which a Facebook profile could be

utilized, thus providing another vast and inexpensive gateway to business users as well,

who could now use the online social networking world to advertise and expand their

businesses. This flexibility and the variety of ways in which a Facebook profile could be

optimized were not just limited to business users. (Kincaid, 2011)

16

2.2 Zero.facebook.com and Mobile Operators

Undoubtedly, Facebook is keeping its firm grasp in the developed Internet

markets such as US and Europe. However, Facebook's collaborative strategy is far-

sighted and equally impactful in emerging Internet markets where it collaborates with

mobile operators in a quest to identify and associate a new mobile internet user's first

encounter with the world of mobile data connectivity exclusively with his / her first

Facebook experience.

The case study of zero.facebook.com is an example of application-specific mobile

Internet traffic discrimination, and a subtle way of traffic prioritization. It represents a

scenario in which Facebook, which is an Over the Top (OTT) application provider, is

collaborating with existing and emerging mobile operators and facilitating them to

expand their subscriber base. Primarily, this is an initiative taken by Facebook as an idea

sold to mobile operators. The idea stems out of Facebook's intention to become the

symbol of Internet for the new social networking application users as they join the mobile

Internet experience.

Two of the significant challenges associated with the mobile Internet usage are

slow data transfers and costly & complex data plans. These factors can deter the users

from using the mobile Internet frequently and seamlessly. Facebook designed

zero.facebook.com to help solve these two barriers with the hope that even more people

will discover the mobile Internet lured by Facebook's brand appeal, recognition, and the

need of association with it. With its already established worldwide popularity, Facebook

desires to penetrate emerging digital markets such as Africa, and wants people to

associate Facebook as the first and possibly the only application intertwined with the

experience of accessing Internet from their basic mobile phones using low cost monthly

plans. Zero.facebook.com is the stripped down version of the social networking

application. It is WAP based, hence lightweight and doesn't consume a large amount of

data bandwidth. The access to zero.facebook.com is free i.e. it does not get charged

17

against the mobile Internet data traffic quota. The incentive for Facebook is that the user

will bond with Facebook in a unique way, as they use this application and identify the

Internet experience through Facebook. The incentive for the mobile operator is to get new

subscriptions and introduce the mobile data services to these new users or the existing

customers. In exchange, the mobile operator will charge for provision of data

connectivity required for accessing and uploading pictures, and non-Facebook Internet

access requirements once the user gets accustomed to the Internet accessibility through

mobile.

The zero.facebook.com site is only available on the networks of 50 operators that

Facebook has partnered with. People can still access Facebook from the standard mobile

site m.facebook.com or Facebook mobile site for touch screen mobile devices,

touch.facebook.com, under their operator's standard data charges.

Some of the operators offering ZERO.facebook.com are:

e Telstra in Australia

0 Movistar in El Salvador

e SFR in France

e XL in Indonesia

- DiGi in Malaysia

e Telecom NZ in New Zealand

e SMART in the Philippines

e Vodafone in Qatar

e Digicel in Suriname

0 Trinidad and Tobago

*3 in the UK

e MTN in Cameroon and Guinea Conakry.

18

2.3 Causal Loop Diagram for zero.facebook.com

WNW" EMMA FAAMO 4

Figure 2-1: Causal loop diagram for Zero.facebook.com

This causal loop diagram primarily depicts and captures the reinforcing loops, which are

benefitting both the Mobile Network Operator (MNO) and Facebook.

The arrows represent the causal relationship between two variables. The arrowhead

points towards the variable affected by the cause, which is attached to the origin of the

arrow. The +ve sign shows that there is a positive correlation or a reinforcing correlation

between the two variables. For example, there is a positive arrow connecting the new

mobile subscriptions to the MNO profitability (Mobile Network Profitability). This

positive correlation implies that when the new mobile subscriptions increase, the MNO

profitability increases. The reason MNO profitability is shown in a rectangular box is that

it is considered as a stock/reservoir variable here. In terms of Systems Dynamics, when a

variable is considered as a stock, there are inflows increasing its value and outflows

decreasing its value. There are two primary loops in this diagram; the exterior loop

representing the Facebook profitability and the interior loop representing the Mobile

19

network operator's profitability. Facebook's profitability is dependent upon the Facebook

users connected to Facebook, whereas, MNO's profitability is dependent on the number

of subscribers as well as the subscribers who are paying for the content accessed through

mobile data connectivity. Hence the common profit driving variable for both Facebook

and MNO profitability is New Paid Facebook Mobile Users. This means that as the

number of New Paid Facebook Mobile Users increases, both the Facebook and MNO

profitability increases.

Here the collaborative innovation takes an interesting twist as the investment in

Zero.facebook.com can be considered as two incumbents (Facebook and MNO) creating

a new concept, fathering a new entrant where new mobile subscribers, user experience

control and customer stickiness (further elaborated in Chapter 5) are the significant links

of this causal loop diagram.

2.4 Lessons Learned from Zero.Facebook.com

2.4.1 Value for user

Collaborating with operators from around the world, Facebook developed this new

mobile site with two main attributes:

2.4.1.a Application Alacrity

Zero.facebook.com includes all the key features of our standard mobile site

m.facebook.com. Users can update their status, view their News Feed, like or comment

on posts, send and reply to messages, or write on their friends' Wall just as they do on

Facebook.com. Zero.facebook is WAP based hence lightweight and efficient. Rather than

making photos viewable on zero.facebook.com, the photos are put one click away so they

don't slow down the experience. Users can still view any photos on Facebook if they

want but their regular mobile data fees will apply.

20

2.4.1.b Zero Cost

Facebook has collaborated with mobile operators to ensure that people can access

O.facebook.com without any data charges. Using zero.facebook.com is completely free.

People will only pay for data charges when they view photos or when they leave

zero.facebook.com to browse other mobile sites. When they click to view a photo or

browse another mobile site a notification page will appear to confirm that they will be

charged if they want to leave zero.facebook.com The specific terms of business

agreement remain undisclosed. However it is speculated that Facebook is discreetly

paying some of the mobile operators to offer zero.facebook.com with new subscriptions.

Facebook management has not responded to such speculations.

2.4.2 Value Chain Evolution- Content Delivery Networks

Zero.facebook.com has presented a novel example of content delivery through mobile

data. Facebook customized the application for low mobile network requirements and this

can be a trendsetting example for value chain evolution where the content providers use

the mobile data connectivity in untraditional ways.

The industry concedes that there is real money in mobile, but this means that bad

performance is a more painful problem that hits revenue numbers and corporate

profitability. At the same time, the evolution of mobile behaviors and use cases (for

users, publishers, and advertisers) underscores the need for new ways to overcome

weakness in wireless data network delivery to ensure that content arrival on handsets can

hold users and maintain high engagement rates. Fortunately, some Content Delivery

Network (CDNs) are starting to think about these business problems and have started

offering mobile content acceleration platforms, laying the groundwork for a parallel CDN

technology solely focused on mobile that integrates nicely within existing CDN offerings.

Zero.facebook.com is an example where the user can switch from the zero.facebook.com

21

to the m.facebook.com seamlessly with a content delivery mechanism that presents

integration between the free and paid-for Internet content. Facebook managed to do this

based on the historical success of the viral social effects, but for other content providers

and newcomers, a lot of work is still ahead and many CDNs do not yet have a true mobile

content offering in the market, let alone a mobile acceleration one. (Rayburn, 2012)

Zero.Facebook.com is one of the forerunners in this trend and is benefitting short-term as

well as long-term with this strategic investment. CDN vendors that don't start addressing

these developments in the market and start offering mobile delivery platforms are going

to have a very hard time diversifying their business and growing their revenue over time.

This is the next avenue of the content delivery business and it is developing in real-time

right now. (Rayburn, 2012)

2.4.3 User Experience Control

"The Pied Piper of Hamelin led one hundred children into a cave from which they

never emerged Some 500 million people, of whom about 10 percent are thirteen to

eighteen years old and another 25 percent are eighteen to twenIyvfive years old, are now

marching to the digital pipes of Mark Zuckerberg, who is twenty-six years old I have no

idea where they are marching, and whether they will ever return."

Richard A. Posner, 2010

The number of Facebook users is astronomical. According to Facebook, 1.11

billion people are using the site each month, up from the 1.06 billion reported three

months earlier, thus representing a 23 percent growth from the previous year. Facebook

reports the number of monthly active users to be 665 million, up 26 percent from the

previous year. Each month, 751 million people access Facebook from a mobile device, an

increase of 54 percent compared to the previous year, and an increase of 10 percent from

the last quarter of 2012. (Tam, 2013)

22

As can be inferred from the above numbers, social media, led by Facebook, has

drastically affected the way we communicate. Social media sites have taken the

communications technology frontier to the next level. Our personal and professional lives

have been reshaped and are being transported to another new dimension inside our brains

where the memory is being fundamentally altered (Nicholas Carr, 2010). In their study,

Sparrow, Liu, and Wegner (2011) observed that "we are becoming symbiotic with our

computer tools, growing into interconnected systems that remember less by knowing

information than by knowing where the information can be found". They deem being

connected to Google as essential, so that we can "know what Google knows" rather than

relying on a conventional memory that is quickly dissolving.

Facebook has influenced life in the modem era in ways and levels previously

unknown and unmatched. David Kirkpatrick (2010) in The Facebook Effect, labels

Facebook "a technological powerhouse with unprecedented influence across modem life,

both public and private" (p. 15). Kirkpatrick notes that in conjunction with other social

media vehicles spawned in its aftermath, e.g. Twitter, Bebo, and Mixi, Facebook offers

individual liberation as well as a "safety in numbers" aspect that can potentially affect

social change in a variety of ways, ranging from organizing political movements and

protests a la Columbia (Kirkpatrick, 2010), to connecting the separated family members

and loved ones in a variety of ways; one such example being the aftermath of the

Japanese earthquake and tsunami (Blackburn, 2011). Zero.facebook.com is spearheading

the same idea of shaping the user experience in emerging mobile markets by

associating the zero.facebook.com as the first image of Internet for a mobile user. It is

interesting to note that this would not have been possible without collaborative

innovation with the mobile operators.

23

Chapter 3

Google-Voice and Sprint

"Once we rid ourselves of traditional thinking we can get on with creating the future."

- James Bertrand

This case study serves as an example of collaborative innovation and value chain

evolution where an Over the Top application is embraced and sold by the Mobile

Network operator to create a win-win situation while generating added value to attract

more customers.

While Skype is one of the oldest voice chat apps, having started in 2003 and about

18 million people online at any moment in time., (Drager, 2011). Google Voice has taken

up the innovative route and encroached on U.S. mobile operators' turf, even offering to

port a user's carrier assigned mobile number for a $20 fee for use with Google Voice's

service. But for one carrier, i.e. Sprint, the potential friction turned into a partnership. On

March 20th 2011, Sprint and Google announced deep integration with Google Voice that

allows subscribers to use their Sprint phone number as a Google Voice number to access

the service's features. Features offered included transcribed visual voice mail (manage

voice messages similar to e-mails), call forwarding (calls ring through on subscribers'

cell phone, home phone, office phone, Gmail inbox and so on), custom voice mail

greetings, and competitive international calling rates, among other additional offerings.

Vincent Paquet, senior product manager for Google Voice, described this

initiative in his words as, "We always felt that the ultimate simplicity would be to just use

Google Voice with your mobile phone".

24

Get even more out of your Sprint phone withGoogle Voice.Sprnt customets can na take advantage of the complete set of Google Voice featureswmhout chlwing their nmber.

When you enable Google Voice on your Sprint phone, you'll get:

One Wanbor for all yew phonesU Pick the phones you wait to ing (home, work etc.)when your Spdnt number Is dialed, so youl never miss a caa.

Online VoicemalCheck your messages by phone, email, or on the web and keep it for as long as you'd like. Plus, your messages will beconverted into text and sent to you via text message or email.

Custamlzaton & controlCreate and assign personaized voicemail greetings based on vht's calling you, block uwrented caters, mark telemarketers asspam. and makeinteratonal cats at our very loi rates.

Figure 3-1: Google-Voice & Sprint advert on Google.com & Sprint.com

Anyone could sign up for the free Google Voice service before, but Sprint-Google

voice agreement brought forth a few notable exclusive benefits. First, Sprint simplified

getting on board with Google Voice. Previously, Google Voice users on any carrier

needed to walk through a number of steps to either get a new Google Voice number, port

their existing mobile number, or let Google handle just the voice mail. The results could

be confusing and cumbersome, with friends often collecting multiple phone numbers for

a contact, depending on several factors, including if the Google Voice user had a feature

phone, or used a Google Voice mobile app or Web-optimized site from a smartphone. As

a second benefit, all Google Voice calls originate from the same single number; the one

first issued by Sprint. Third, if you enable Google Voice, the service replaced Sprint's

voice mailbox on your phone, so dialing "1" from the handset dialed up your Google

Voice message inbox. No additional setup on the Internet VoIP application would be

required. Fourth, Sprint smartphones get most Google Voice features without requiring a

mobile app. Mobile texting is one exception to this last point, however, Sprint's rates and

plans still apply for messages sent from the phone's default texting program. Google

Voice texts remain free to the United States and Canada if sent from the Web or from a

Google Voice smartphone application.

25

Prepaid phones and Mobile Virtual Network Operators (MVNOs) that ride on

Sprint's network, like Boost Mobile and Virgin Mobile, are ineligible for the

Sprint/Google Voice combined package.

Similar to Zero.facebook.com, the terms of business agreement between Google

Voice and Sprint remain undisclosed. The possibility of Google Voice partnering with

other mobile operators is not ruled out especially in light of the integration and hand-over

issues between Sprint and Google Voice. One prominent integration issue was that during

international roaming the consumers get charged for Sprint cellular voice minutes instead

of being charged the cheaper Over the Top Google Voice minutes.

FAILEEOliTA11ON

CHURN CW

USm

EXERW~

CONTROL

CAPEXAOCR

NM. m ?MGOCRZ ROFIARILMT

PC"A0UM0VOIPW 41CEILLULAR IDUTES VA IPsusMwo

SHARE

Figure 3-2: Causal loop diagram for Google-Voice and Sprint

The arrows represent the causal relationship between two variables. The arrowhead

points towards the variable affected by the cause, which is attached to the origin of the

arrow. The -ve polarity arrow shown with red color in the Figure 3-2 above originating

from Churn and leading to MNO profitability shows the simple negative causal

26

relationship between these two variable; as chum increases the MNO profitability

decreases. Packaging VoIP with Cellular Minutes increases the Value for Mobile

Subscriber, which in turn increases the New MNO Subscription. As the Consumer

Stickiness increased in this case, so did the Consumer Expectations from the service

offering. The integration failure issues gave rise to Failed Expectations, which increased

the Churn.

3.1 Lessons learned from Google-Voice and Sprint casestudy

3.1.1 Co-opetition

This case study is an effective example of co-opetition (collaboration combined with

competition, or cooperation between competitors) between a mobile service provider and

an Over the Top Voice over IP (VoIP) service provider; Sprint joined hands with an

apparent competitor and jumped ahead of other operators in opening a new co-opetition

avenue with a traditional value chain rival Google Voice. The ease of integration and

price effectiveness provided value for user and increased subscriptions for both Google-

Voice and Sprint. Hence, the first lesson to be extracted out of this case study is that

competitors using different technological makeups can actually act as complimenters.

This same phenomenon is now being observed in other parts of the telecommunications

value chain. With the advent of Long Term Evolution (LTE) technology for cellular

operators and the question of monetization of Quality of Service of data, an integration

between mobile broadband and Wi-Fi hotspot broadband networks, at interface as well as

traffic management policy framework level, for offloading and seamless QoS parameter

handover is highly likely in the near future.

3.1.2 Integration Failures

Integration failures represented by this case study are a good example of possible

negative repercussions of collaborative ventures. Failures of technical interface

27

integration between Google-Voice and Sprint for the international roaming mobile users,

and lack of integration in the texting/SMS (Short Messaging Service) arena, proved to be

the primary reasons for failed user expectations. The collaborative failure rate was

enhanced which reduced the value for user and in turn reduced customer stickiness. It is

also imperative to note that this setback did not affect Google-Voice revenue much

because being the Over the Top platform, the users over the internet and non-Sprint

subscribers could still use Google-Voice and map the Google-Voice number to other

mobile carriers. Another significant factor is that Google itself has other dimensions to

sustain itself and Google-Voice is still experimenting and also willing to share similar

collaborations with other mobile operators. Sprint, however had a setback since they were

already losing revenue on regular mobile subscriptions and voice minutes. The mobile

operators are expected to provide seamless support over their network and the value

added services or features that they introduce should be covered end to end to cater for

special customer needs with the traditional operator services such as international

roaming. This observation is endorsed by the aggregated model, presented in chapter 5

and supported by the results depicted in Figure 5-1 & Figure 5-2 (page 43- 44 of this

thesis).

3.2.3 Negative Feedback from User Experience Control

The Google Voice- Sprint venture's most prominent selling point was its price

sensitivity. The international roaming integration failures caused the consumers to pay

more because they were forced to use the Sprint mobile minutes instead of low-priced

Google-Voice minutes while travelling overseas. This reduced the value for user and

magnified the sense of entrapment that is incorporated in the perception of intrusiveness

parameter in the final aggregated model presented in Chapter 5. While a platform offers

benefits of exclusivity, it can also backfire, if the users are unable to enjoy the benefits of

other free services available in the market. A similar setback was witnessed in the case

study of NTT DoCoMo i-mode, discussed in the next chapter.

28

Chapter 4

NTT DoCoMo i-mode

"Learning faster than your competitors is the only sustainable competitive advantage in

an environment of rapid change and innovation."

- Arie de Geus, 1998.

In this case study, the years under focus are between 1999 and 2007, which were

the peak years of NTT DoCoMo i-mode's success and subsequent setbacks. In 1999, the

Japanese population numbered 126 million. Of this 126 million, only 12.2 percent of the

population had Internet access, compared with 39 percent of the US population, 21

percent of the British population, and 23 percent of the Korean population. In a study

conducted by AOL and Roper ASW, 69 percent of Japan's online population said the

Internet was essential to everyday life, but 29 percent said that dial-up telephony costs

were the biggest obstacle to Internet access (Mullins, 2007).

Despite the relatively low prevalence and popularity of Internet usage, the mobile

handset usage and popularity was far ahead in Japan as compared to other industrially

developed countries. At the end of 1999, 44.5 percent of the Japanese population had

mobile phones, compared with 40 percent in the UK, and 31 percent in the USA. Dial-up

telephone access was expensive in Japan and consumers were obsessed with media and

information access; information ranging from daily stock exchange rates to weather

updates to latest comic publications. These facts presented Japanese market as an ideal

stage to sell a reasonably priced mobile data service targeted towards specified segments

of users with the content and information of their choice, available on their mobile

handsets 24/7 (Mullins, 2007).

29

i-mode is a mobile Internet service that caused a revolution in both business and

private lifestyles in Japan. Despite the uniqueness of Japanese regulatory controls and the

monopolistic situation NTT DoCoMo enjoyed, they improvised as an operator to jump

ahead in the value chain by providing content availability and Internet experience to

customers in a unique fashion.

By 1990, the Japanese mobile market had enjoyed meteoritic growth at a pace

unmatched by any other country in the world. In the latter half of 1990, the Japanese

mobile market was on the verge of reaching maturity, even though not complete

saturation, when NTT DoCoMo developed a novel service in the form of an innovative

mobile Internet platform with the aim of promoting a further evolution in mobile

communications. The i-mode service was launched in 1999 attracting overwhelming

support from mobile phone users. i-mode not only created new profitability in the mature

mobile phone market, but also redefined mobile communications for the new age by

providing users with an incomparable service.

i-mode by NTT DoCoMo Japan is a walled garden mobile Internet interface in

contrast to the traditional Internet access via mobile handset. NTT DoCoMo's i-mode

proved to be a revolutionary service in Japan. i-mode offered its users a wide variety of

services, including web access, e-mail and the packet-switched network that delivers the

data. i-mode users have access to various services such as e-mail, sports results, weather

forecast, games, financial services and ticket booking through a customized interface

called i-Menu. Content is provided by specialized services, typically from the mobile

carrier, which allows them to have tighter control over billing.

Similar to zero.facebook.com that uses WAP, i-mode delivers only those services

that are specifically converted for the service, or are converted through gateways. This

has placed both systems at a disadvantage against handsets that use "real" browser

software, which used a flat pricing structure for data. Even i-mode's creator, Takeshi

Natsuno, has stated, "I believe the iPhone (a phone that uses the traditional TCP/IP

30

model) is closer to the mobile phone of the future, compared with the latest Japanese

mobile phones."

During the i-mode case study years under consideration in this thesis (1999 to

2007) i-mode had been extraordinarily successful in Japan. This was because of its

outstanding convenience and its unique business model. Its reliance on this system offers

an innovative approach to the mobile service value chain and to wireless service / Internet

relationships.

Ingenuity is exemplified in the i-mode business model. It synchronizes all value

chain aspects, ensuring that content, quality, wireless technology collectively driving the

user experience control. The billing system is streamlined by bill consolidation, with

DoCoMo also collecting information access fees on behalf of i-Menu-listed content

providers.

1-mode business model

Figure 4-1: i-mode business model, Source: NTT DoCoMo website

i-mode collaborated closely with equipment manufacturers, content providers, and

other platforms to ensure that wireless technology, content quality, and user experience

evolve jointly. This synchronization guarantees that customers, partners and shareholders

share interests with end-users, thus enabling all parties to maximize value and to continue

to improve the quality of products and services connected with i-mode. i-mode's adoption

of HTML subset made the creation of sites simple in addition to offering compatibility

31

of many other standards, including GIF, Java, MIDI, Macromedia Flash, and HTTP. i-

mode upholds the available quality content through the i-Menu. Fig 4-2 shows evolution

of these mobile web standards. All content is continually updated, enhanced and designed

to support clarity and application. i-mode used the provision of high-quality content is

used as a means of attracting customers.

Evolution of Mobile Web-Related Markup Languages

DPW-

11sn% 114

. ........

1 ~ ~ -1 N l Iv1 i

1M IM 162 1993 10 1s 1m 17 IM IM 20M 201

,:I :r

Figure 4-2: Evolution of mobile web standards -M. Stuckwisch(2007)

In contrast with the Wireless Application Protocol (WAP) standard used for

zero.facebook.com, which used Wireless Markup Language (WML) on top of a protocol

stack for wireless handheld devices, i-mode utilizes fixed Internet data formats such as C-

HTML based on HTML, as well as DoCoMo proprietary protocols ALP (HTTP) and

TLP (TCP, UDP).

i-mode phones had a special i-mode button for the user to access the start menu. There

were more than 12,000 official sites and around 100,000 or more unofficial i-mode sites,

32

M0220 20M 2 00 M 200

which were not linked to DoCoMo's i-mode portal page and DoCoMo's billing services.

NTT DoCoMo supervised the content and operations of all official i-mode sites, most of

which were commercial. These official sites were accessed through DoCoMo's i-mode

menu but in many cases official sites can also be accessed from mobile phones by typing

the URL or through the use of QR code (a barcode).

NTT DoCoMo authorized all i-Menu content, while quality was maintained by

setting high operability standards and offering quality services. Other content providers

complement these services via their own sites as demand dictates. The i-mode service

was thus energized by attracting more subscribers, enabling added high-quality content.

The operator NTT DoCoMo controlled the user pricing and billing by collecting

monthly information charges for the i-Menu listed content providers via a consolidated

bill for all mobile phone activities, thus eliminating the need for provider billing. This

arrangement reduced expenses for the content partners and encourages them to generate

high-quality offerings to attract new subscribers, thereby boosting their profits.

Additionally, NTT DoCoMo was able to generate incremental revenue by charging a

small commission for the clearinghouse billing system service. Peak profit years were

between 1999 to 2006; with over 22 million subscribers within the first 2 years. The

service currently boasts of more than 95,000 Internet sites providing a variety of

content. By the middle of 2001, within a short span of two years, Japan's wireless phone

service i-mode had signed up nearly 20 per cent of the total Japanese population, or 25

percent of the population between the ages of 15 and 64, and became the mostly widely

used mobile Internet service in the world.

33

4.1 NTT-DoCoMo i-mode Causal Loop Diagram

DEVICE + IMODE +COMMISION FOR INCOPATIBILITY PROFITABILITY

CONTENT PROVISION

CUSTOMERSTICKINESS

VOICE & TEXTINVESTMENT IN REVENUE

DIFFERENTIATING +

USER SERVICESEXPERIENCE

SUBSCRIPTION±TO i-MODE

BILLING OVERHEADFOR CONTENT VALUE FOR USER

PROVIDERS VBETTER CONTENT

QUALITY +

DEMAND FORCONTENT CONTENT

"DEVELOPMENT

+ CAPITAL ~-~~---b +AVAILAL

CONTENT VARIETY+

Figure 4-3: NTT DoCoMo i-mode causal loop diagram

The Device Incompatibility has a negative relationship with Customer Stickiness, whereas

the User Experience Control is positively correlated with Customer Stickiness. Available

Content Variety and Better Content Quality drive an increase in the Value for User which

in turn increases the Subscription to i-mode. In this diagram the i-mode profitability is

represented as a variable whereas the content provider is gaining the benefit of

collaboration through the increase in Content Development Capital.

34

4.2 Lessons Learned from i-mode

4.2.1 Shaping User Experience Control

NTT DoCoMo's i-mode presents an ideal example of how an organization can control

and shape the user experience. This case study presents a vivid and thorough example of

controlling and shaping user experience by using technology and collaborative

innovation. A noteworthy fact of this case study is that in addition to enjoying the

traditional role of a mobile operator whose association with the end user is based on the

billing relationship, NTT DoCoMo further enhanced this privilege into a consumer

shaping experience mechanism by collaborating with different players in the value chain,

specially the device manufacturers and the content developers. The range of this

integration gave NTT DoCoMo a unique advantage and unrivaled control over shaping

the user experience after determining the user's needs.

One of the primary reasons of DoCoMo's new service being an instant hit was because of

its designers' intimate familiarity with the Japanese market on a segment-by-segment

basis. The existing customer base and profiling advantage that NTT DoCoMo enjoyed in

the Japanese market helped it in identifying the potential customers for targeted services

based on the content and information salient to the respective segment of the market.

DoCoMo segmented its markets behaviorally and designed offerings of different

downloadable information for each segment: those interested in financial markets, comic

strips, cartoons and so on.

4.2.2 Value for User

Characterized by the traditional slow transmission speed of the mobile Internet access in

the late '90s, i-mode's successful strategy included creating more value for the user by

charging the user not based on the time spent on the Internet

(unlike the traditional wireline Internet Service Providers (ISPs). Users were charged

based on the amount of information downloaded and not the connection time. Emails cost

1 yen per 20 Japanese characters (40 roman letters). Downloading still images cost 7 yen,

35

checking share prices cost 26 yen, and transferring funds from bank accounts cost 60 yen.

While some of i-mode's content providers charged a flat monthly fee, others were free of

charge. In 1999, i-mode charged a basic monthly fee of 300 yen ($3.5) and a packet fee

(based on the volume of data sent or received) of 0.3 yen per 128 bytes of information. i-

mode also priced the mobile phone handsets reasonably in comparison to other mobile

phones available in the market.

An i-mode user paid for both sent and received data. Unsolicited emails could be avoided

through the email service hence making the cost fair. The basic monthly charge was

typically on the order of JPY Y200 - Y300 for i-mode not including the data transfer

charges, with additional charges on a monthly subscription basis for premium services. A

variety of discount plans were also offered, for example family discount and flat packet

plans for unlimited transfer of data at a fixed monthly charge (on the order of

V4,000/month).

In addition to low price, the billing method was also convenient for users. Instead of

paying i-mode for service fees and paying the content providers for subscription fees, i-

mode customers received one monthly bill with all of their mobile charges. "The i-mode

system has made m-commerce [mobile commerce] a reality in Japan by introducing

information billing systems that attach charges directly onto telephone bills", said

Natsuno.

4.2.3 Collaborative Innovation based on user needs

One target market that intrigued Takeshi Natsuno, Executive Director of NTT DoCoMo,

included consumers interested in the financial markets and their own personal finances.

To appeal to this group, i-mode developed relationships with the banking industry. "Of

the more than 700 content partners we have, 320 are banks", said Natsuno.

Another target market comprised customers with an eye for comics. To serve this

segment, i-mode contracted the publishing firm Shueisha to provide weekly comic strips

36

for a monthly fee of 300 yen (less than $4) for the transmission of a weekly comic strip.

The toy company Bandai sold charappa or cartoon characters. For less than $2 a month,

subscribers received a different cartoon image on their phone every day. By February

2000, Bandai had 400,000 i-mode subscribers.

As Natsuno said, "The success of i-mode is because we adjust our site to Internet users."

Furthermore, unlike a dial-up Internet connection, i-mode Web access was always on,

allowing customers to use the Internet without dialing the phone. Even the phones were

appealing to the Japanese market, with color screens, lightweight handsets, multi-link

navigation and better graphics capabilities. The only disadvantage of the product was its

transmission speed of 9.6 kilobytes per second (Mullins, 2007).

4.2.4 Win-Win arrangement with Content Providers

NTT DoCoMo's insight into the needs of its content providers was an important

contributor to its early success. By taking care of the customer billing, i-mode made

business easy for content providers, who were hesitant to sell online because handling the

billing was a large and expensive burden. By paying NTT DoCoMo to do the billing, the

content providers were able to concentrate on what they did best - providing content -

and still generate earnings. In return for this, i-mode charged its content providers 9 per

cent of revenue.

The company also kept a firm grip on its business, controlling all aspects of the i-mode

service. Unlike some European promoters of wireless application protocol (WAP),

another mobile data technology, DoCoMo knew that developing content would be

crucial. DoCoMo required its content providers to create wireless content from the

ground up, specifically generating content to fit the mobile phone format. DoCoMo's

success lay not in its technology, which actually was not state-of-the-art, but in its ability

to bring together and control all these pieces and thereby deliver content that its target

customers wanted.

NTT DoCoMo created i-mode at a time when the Japanese market for mobile phones was

37

reaching maturity and users were in need of new services. Its foresight and customer

understanding led to impressive results.

The i-mode story also provides an example of the frequently raised questions of whether

the technology should drive business ventures, or the consumer experience improvement

and value for the end user should drive the business ventures/collaborations and this

thesis argues for the latter. In i-mode's case, Internet and communications technology

created the possibility of delivering information to mobile customers, any time,

anywhere. Thus it is evident that once there is a valid consumer need, the business need is

generated there and then, and the supporting technology can be either developed in the

form of firm innovation or collaborative innovation by synergizing technological

capabilities across the value chain or value network. However, it is imperative to have a

clear target market, a clear customer need, and that what the opportunity offers satisfies

that target market's need in a way that's faster, better, cheaper or otherwise more

beneficial - generating more value for the user and hence elevating the customer

stickiness and driving user experience control.

4.2.5 Global Trendsetter

A few months after DoCoMo launched i-mode in February 1999, DoCoMo's competitors

launched very similar mobile data services: KDDI launched EZweb, and J-Phone

launched J-Sky. Vodafone later acquired J-Phone including J-Sky, renaming the service

Vodafone live!, although initially this was different from Vodafone live! in Europe and

other markets. In addition, Vodafone KK was acquired by SoftBank, an operator of

Yahoo! Japan in October, 2006 and changed the name to SoftBank Mobile. As of June

2006, the mobile data services KDDI, EZweb, and J-Sky, had over 80 million subscribers

in Japan.

Seeing the tremendous success of i-mode in Japan, many operators in Europe, Asia and

Australia sought to license the service through a partnership with DoCoMo. Takeshi

Natsuno was behind the expansion of i-mode to 17 countries worldwide. Kamel Maamria

38

who was a partner with the Boston Consulting Group and who was supporting Mr.

Natsuno, is also thought to have had a major role in the expansion of the first Japanese

service ever outside of Japan.

i-mode showed very fast take-up in the various countries where it was launched which

led to more operators seeking to launch i-mode in their markets with the footprint

reaching a total of 17 markets worldwide. i-mode was launched in the following

countries:

" Australia, (Telstra)

" Belgium (Base)

- Bulgaria (Globul)

- France (Bouygues Telecom (company))

- Germany (E-Plus)

" Greece (Cosmote)

" Hong Kong (3)

- Israel (Cellcom)

- Ireland (02)

" Italy (Wind)

" Netherlands (KPN)

- Russia (MTS)

" Romania (Cosmote Romania)

" Singapore (StarHub)

" Spain (Telef6nica)

" Taiwan (Far East Tone)

- UK (02)

4.2.6 Point of Failure

While the i-mode service was an exceptional service that positioned DoCoMo as the

global leader in value add services, another key success factor for i-mode was the

Japanese smartphone makers who developed state of the art handsets to support i-mode.

As i-mode was exported to the rest of the world, Nokia and other major handset vendors

39

who controlled the markets at the time, refused at first to support i-mode by developing

handsets which support the i-mode service. The operators who decided to launch i-mode

had to rely on Japanese vendors who had no experience in international markets. As i-

mode showed success in these markets, some vendors started customizing some of their

handsets to support i-mode, however, the support was only partial and came late in time.

While the service was successful during the first years after launch, the lack of adequate

handsets and the emergence of new handsets from new vendors, which supported new

Internet services (Oomori, 2008) on one hand, and a change of leadership of i-mode in

DoCoMo, lead to a number of operators to migrate or integrate i-mode into new mobile

Internet services.

40

Chapter 5

The Aggregated Model

Based on the lessons learned from the three case studies described in the previous

chapters 2,3 and 4 and the individual causal loop diagrams (Figure 3-1, Figure 4-2 &

Figure 5-4), an aggregated model was developed as illustrated in Figure 5-1 and Figure 5-

2 below. Figure 5-1 depicts the causal loop structures while Figure 5-2 uses shadow

variables for the ease of understanding the parameter interconnections and dependencies.

Following are some of the important parameters used in the aggregated model.

e FIRM INNOVATION

In the aggregated model presented in this thesis, firm innovation can be described

as the innovation that a firm develops in-house. This could be measured in the

number of inventions, patents or even the new products launched by a firm.

However in this model, "service features" are used as the units to measure firm

innovation.

- COLLABORATIVE INNOVATION

Collaborative innovation is the innovation where two or more organizations

synergize their resources to either develop a new technology or launch a new

service. Similar to "firm innovation", "collaborative innovation" is measured in

units of service features in this thesis.

- USER EXPERIENCE CONTROL

41

"User experience control" is a parameter in this aggregated model, which

describes the service provider's capability to own and shape a user's experience.

Other factors such as consumer profiling and customization of service offerings

based on consumer needs collectively work to enhance user experience control.

Traditionally it was believed that billing Interface ownership was the primary

criteria for being closely aligned and associated with the customer. However, this

belief has been challenged repeatedly in the recent years. The results from the

model used in this thesis also endorse the fact that billing interface ownership

might delay the customer churn by some brief time period, but would not ensure

customer loyalty in the long run, by itself.

PERCEPTION OF INTRUSIVENESS

As obvious from the name, this parameter refers to he consumer's feeling of being

intruded upon, which is not only based on the loss of privacy (a possibility in the

case of consumer profiling and data mining which probes into the user's

consumption of applications and choice of content etc), but also on the user's

perception of confinement or entrapment because of the closeness/inflexibility of

a certain platform or service and the lack of integration opportunities. One such

example is the lack of compatibility of i-mode handsets with the open Internet and

the unavailability of free content over the i-mode interface.

Rest of the parameters are described in the section 5.1 below:

42

------ -------

RMLAM

PAOFFPJIRM

As 3

FigureVX 5-+ The Ag rgtd oe it p

43

. ....................... ... .. -- ...... .....

Barriers to Newd wcuer d s e . Entrants

Firm In tion RateProfitability ofService Provider Customer

Word of Mouth maintenane O irnnIovatio firm innovationfailure rate

Revenue

Adoption rate ARPU r4- s of servc Competiiv Advantago

Total Customer

cutmrstickiness obsoletion rate

+ + Consumer spending Collaborative =(on QoS e claborstive Innovation

Existing customers Pretono au Fai sr ra e gainf

Patform Enrichment innovation rateBilling Interface -- -

/ + Ownerships pfilngx profitability Lookup e fft of capi on

+User + ma cpialinovtion rat

Experence User ex lbaialprsevcUser Experience C control fereControl Lons rate fea101

r c c Normal innovation rate

service feature

Figure 5-2: The Aggregated Model with Shadow variables to improve visibility of

parameter dependence. (Shadow is a copy of a system dynamics variable used to reduce the

complexity of the diagram and overlapping of arrows)

5.1 Parameters Evaluated In The Model

The following explains the causal dependencies. When it is said that A is a

function of B , it implies that changes in A affect B.

Following is a list of the variables /parameters captured in the aggregated

model presented in Figure 5-1 and Figure 5-2 above:

(01) Adoption rate is a function of: Perception of Value for Money as well as Word of

Mouth.

(02) ARPU (Average Revenue per User) is a function of:

Consumer spending (on QoS etc) as well as Platform Enrichment

44

.... . .. ... ...... .

(03) Barriers to New Entrants is a function of:

Customer stickiness as well as change adaptability by user. This variable can be

quantified with considering the other parameters e.g. the Firm innovation of one

player in the value chain.

(04) Billing Interface Ownership is a function of: Firm Innovation

(05) Capital per service feature is a function of:

Maximum capital, which is governed by the profitability of the service

provider

(06) Change adaptability by user is a function of:

Total Customer base, a positive causal relationship with Perception

of Intrusiveness; it has a negative or balancing relationship to User Experience

Control

(07) Collaboration Failure rate is a function of:

Innovation obsoletion rate as well as integration faults

(08) Collaborative Innovation is a stock or level

Collaborative innovation rate and Collaboration Failure rate

(09) Collaborative innovation rate is constant

(10) Competitive Advantage is a function of: Collaborative Innovation as well as Firm

Innovation-competitors average innovation level

(11) Competitors Average innovation level is a function of:

Units: service features

(12) Consumer profiling is a function of: Customer stickiness

45

(13) "Consumer spending (on QoS etc)" is a function of: Perception of Value for

Money

(14) Cost of Service Provider is a function of: Billing Interface Ownership, Total

Customer base, Time, Firm Innovation Rate, collaborative innovation rate, Total

Customer base and Customer maintenance OPEX

(15) Customer maintenance OPEX is the operational cost of maintaining one customer

per month

(16) Customer stickiness is a function of: Barriers to New Entrants, Perception of

Value for Money and Existing customers

(17) Effect of capital on innovation rate is a function of Profitability Lookup, capital

per service feature, reference capital per service feature, and Profitability of

Service Provider

(18) Existing customers is a constant, which was given values between 1000 to 20

million for different classes of service providers

(19) FINAL TIME is the final time for the simulation = 2053

Units: year

(20) Firm Innovation is a level or stock with inflow as Firm Innovation Rate and

outflow as firm innovation failure rate.

Units: service features

(21) Firm innovation failure rate is a constant and varying values were utilized for test

purposes depending upon the technology and its lifecycle.

46

(22) Firm Innovation Rate is a function of: innovation rate as well as collaborative

Innovation

(23) Initial time is the initial year of stimulation = 2013

(24) Innovation obsoletion rate is the rate at which innovations or new features become

obsolete with time. This obsoletion can be caused by the outgrowth or fall-out of

any one of the links completing the value chain. e.g. in the case of NTT DoCoMo

i-mode, the i-mode handset technology obsoletion rate was driven by the newer,

faster, cheaper and more openly interfaced mobile handsets in the market.

(25) Innovation rate is a function of: Normal innovation rate and effect of capital on

innovation rate

(26) Integration faults was variably tested as a constant as well as a function of

collaborative innovation service features.

(27) Maximum capital was tested as a constant and as a function of profitability

(29) Perception of Intrusiveness is a function of: Consumer profiling as well as User

Experience Control

(30) Perception of Value for Money is a function of: Competitive Advantage and

Platform Enrichment

(31) Platform Enrichment is a function of: Collaborative Innovation as well as Firm

Innovation

(33) Profitability of Service Provider is Level or stock determined by Revenue-Cost of

Service Provider.

47

(38) Total Customer base is a stock, with inflows being Adoption rate as well as

customer stickiness with customer stickiness also being a function of existing

customer base and outflow as change adaptability by user.

(39) User Experience Control is a level/stock with the User Experience Control rate as

an inflow, and User Experience Control Loss rate as an outflow.

(40) User Experience Control Loss rate is a function of: Perception of Intrusiveness-

Perception of Value for Money.

(41) User Experience Control rate is a function of: Billing Interface Ownership,

Consumer Profiling and Platform Enrichment.

(42) Word of Mouth is a function of: customer stickiness

Specific equations for the

Appendix C of this thesis

parameters used in the aggregated model can be found in

5.2 Intuitions endorsed by the model

5.2.1 Collaborative Innovation

"When all think alike, then no one is thinking."

- Walter Lippman

As Christensen, in his book "The Innovator's Prescription" writes "When

disruptive innovators attempt to commercialize their innovations within the established

value network in their industry-essentially trying to cram it into the back plane of

48

competition-that system will either reject it (as it did with Sony's transistorized

products) or co-opt the potential disruption, forcing it to conform to the existing value

network in order to survive".

Christensen further says in his book "Creating an appropriate business model is

essential to making disruptive innovations successful. And creating an appropriate value

network is critical to making disruptive business models successful."

As discussed earlier in Chapter 1 of this thesis, while presenting the evolution of

mobile value networks, it is imperative to develop the straightforward value chain into a

value network in order to break the norms and collaborate for a service which

amalgamates two different dimensions of the existing value chain e.g. content developer

and mobile operator or device manufacture and infrastructure provider.

Following are some of the intuitions endorsed by the aggregated model:

* For a value chain player with an existing loyal customer base, the collaborative

innovation failure rate affected the profitability only where the platform

enrichment offered via firm innovation was not sustainable. This is aligned with

the observations in the Google-Voice and Sprint case study where International

Roaming integration failure was a major cause of customer churn. The NTT

DoCoMo' s integration failures with newer models of handsets proved to be a

point of failure in the integrated value chain that NTT DoCoMo had extended for

the i-mode platform.

- Depending on the initial customer base and the resulting impact of firm

innovation failure or success rates on consumer stickiness, the collaborative

innovation can aggravate or alleviate barriers for new entrants. Figure 5-3 and

Figure 5-4 show the impact of firm innovation and collaborative innovation on

49

barriers to new entrants. In Scenario 1, the collaborative innovation increases firm

innovation. In Scenario 2, the collaborative innovation increases similar to

Scenario 1, but firm innovation keeps declining with time. As depicted in the

Scenario 1 and Scenario 2 in Figure 5-3 and Figure 5-4 respectively, the barriers

to new entrants are amplified if the firm innovation increases at some rate over the

years (collaborative innovation also adds to firm innovation by increasing the

knowledge base of an organization). However, if the collaborative innovation fails

to impact firm innovation as depicted in Scenario 2, and the firm innovation fails

to grow independently as well, then the barriers to new entrants are reduced.

Billing Interface ownership can only have a delay in consumer adaptability to

change. In the long run, it is the perception of value for money, which ensures

consumer loyalty.

80

6COLLABORATIVEINNOVATION

inFIRM INNOVATION40

30 - -- Barriers to NewEntrants-A

20

10

0FM 0-3: nari M -Bar 0 n e a

r..4-~ '-4N N N N N

Figure 5-3: Scenario 1 -Barriers to new entrants- via consumer stickiness

50

90

80

70

60

50

40

30

20

10

0

- COLLABORATIVEINNOVATION

FIRM INNOVATION

Barriers to NewEntrants-B

--

Figure 5-4: Scenario 2- Barriers to new entrants- via consumer stickiness

5.2.2 User Experience Control

According to Eyre (2008) "mobile is not just portable online, it is the first truly

spontaneous medium, enabling interactions of an entirely different kind to the computer".

He argued that combining social network type of information with GPS(Global

Positioning System) will enable mobile and social contacts to become more real than

virtual. He also stated that popular, location based and personalized new content makes

way for new advertising opportunities.

The use of location based services for targeted advertisements and mobile

marketing is an existing example of User Experience Control. Mobile phones personify

personal experience and the mobile operators must align themselves closely to the mobile

user's needs in order to keep their customers satisfied. Previously, it was believed that the

service provider who provides the customer with the monthly bill statement and holds the

address and credit card/other payment details of the consumer has a closer relationship to

51

the customer. However, as the mobile and internet industries have become more

competitive, customers expect more value for their money since they have more choice

and they are not as bound to their monthly billed mobile connection as they used to be

with a wireline home telephone because of the physical copper connection restraints.

Mobile technology itself is more personalized, as the handset is smaller and mobile and

usually a user's very personal and handy tool for communicating with the rest of the

world. The plethora of applications available for all practical purposes, ranging from

tracking the public transport, to checking the local and international news to streaming

content and taking, storing and exchanging photographs, have made the mobile

experience more personal than it ever was. These applications range from necessity-based

such as Global Positioning System (GPS) tracking to entertainment-based such as

streaming video content, and play a vital role in shaping a consumer's experience and

represented by the user experience control variable.

PLATFORMENRICHMENT

PERCEPTION FORVALUE FOR MONEY

USER EXPERIENCECONTROL

PERCEPTION OFINTRUSIVENESS

Figure 5-5: Graphical representation of Platform Enrichment, Perception of Value for

Money, User Experience Control, Perception of Intrusiveness over a period of 20 years

- As depicted in Figure 5-5, there is a threshold point of perception of intrusiveness,

which closely follows the increase in user experience control. Beyond this

52

100

90

80

70

60

50 - -

40

30

20

10

0

threshold point, the perception of value for the user starts declining despite the

constant rise in platform enrichment.

- Not just the firm innovation, but the collaborative innovation can also cease to

reap its benefits if the user's "perception of intrusiveness" as well as "entrapment"

which originates from lack of openness of the platform rises with time. This

phenomenon was also witnessed in case of Google-Voice and Sprint where the

traditional benefits of Over the Top voice connectivity's price sensitivity could no

longer be enjoyed by users since they were restricted to the use of Sprint mobile

connectivity minutes instead of the cheaper Google-Voice VoIP (Voice Over IP)

minutes while travelling abroad and while being connected via international

roaming. Similarly, the lack of "openness" and integration with newer handsets

and the lack of free access to Internet content became a major cause of demise in

NTT DoCoMo i-mode's value for end user.

- Another important observation resulting from the evaluation of the aggregated

model is the possibility of using this model for two or more players of the

telecommunications value chain simultaneously. In this thesis, the model operates

for one value chain player. Further work can be done to amalgamate the layers

representing different players in the value chain to gauge their actual individual

share in the available consumer surplus (the maximum amount of money a

consumer or set of consumers are willing to spend on a certain service).

53

Chapter 6

Rationale Behind Monetization of QoS

"The pipe cannot be just dumb in times of capacity constraints. The smart pipe is the

future... with different levels of service connectivity and the ability to manage traffic."

Rene Obermann, CEO Deutsche Telekom,

Mobile World Congress, Barcelona, 2012.

Sharing the same stage with Obermann, the fellow keynote speaker Ben Verwaayen,

CEO of Alcatel Lucent, was a little more forthright when he declared, "I am stunned by

the fact that we have to stay non-discriminatory in this industry; it's ridiculous." He

added, "This market will have to develop a form of choice for the consumer on quality.

Without it, this market can simply not do all the things we need it to."

In the previous chapters, the three case studies presented examples of telecom

operators collaborating with other players in the value chain through more of business

cooperation and less of technical joint ventures to generate revenue. The monetization of

QoS of data is a technical capability, which can be used by the mobile operators to

enhance revenue as well as play a more important role in shaping user experience control

and evolution from a value chain into a value network (further explained in section 6.3 pg

59 of this chapter)

For the last decade, and particularly over the last few years, the communications

ecosystem has evolved quite rapidly. Most of the interviews with stakeholders in the

telecommunications industry resonated with the common perception from the network

operators, both fixed and wireless, that over the top service providers have been claiming

54

the lion's share of profitability while network providers, have been haunted by the ever-

increasing fear of being reduced to dumb pipes. While often viewed as a threat, this set of

circumstances can be viewed as an opportunity for cooperative innovation where network

operators provide QoS for mobile data services, with the user controlling which

applications will get priority.

6.1 Mobile Data: Imbalance in demand and capacity

The concept of mobility has transformed the telecommunications industry within the last

decade. It has changed the social fabric of human civilization and opened new avenues of

business and technology. Close to 83 percent of the world's population now enjoys

access to a mobile phone. In 97 countries around the world, there are now more mobile

devices than people (ITU, 2010). The drastic improvements in mobile handsets including

smartphones such as iPhones, Android-based smartphones, and tablets, and the plethora

of content and application availability have further increased the demand and usage of

mobile Internet data. Courtesy of the attractive social networking sites, gaming websites,

video streaming and ground-breaking usage of Internet for business promotion and

business management, the mobile networks have seen an overwhelming demand of data

connectivity specially in the last couple of years. A market research by the Cisco@

Internet Business Solutions Group (IBSG) revealed that more than three-quarters of

Americans were watching videos, updating their social networks, and participating in

video conferences on their mobile phones by late 2012 (Cisco IBSG,201 1). Mobile data

connectivity ensures that people are connected to their email, their favorite applications

and social networking sites anywhere, anytime.

The obvious result of this demand of mobile connectivity is an astounding growth in

mobile network traffic. Cisco's Visual Networking Index (VNI) predicted that global

mobile data traffic will increase 26-fold between 2010 and 2015, reaching 6.3 exa bytes

per month by 2015. Global mobile traffic will continue to explode, growing at a rate three

times faster than that of fixed IP traffic over this same period. (Cisco VNI, 2011)

55

The fierce competition, price wars, lost revenue in voice minutes, and the fast and furious

innovations in mobile handheld smart devices forced the mobile operators to compete by

promoting subsidized data usage offerings bundled with these data-hungry devices. A

very relevant example is the iPhone promotional offerings by AT&T and many other

carriers, specifically in the US, and generally around the world. The evolution of the

telecommunications ecosystem compelled mobile operators to support data-hungry

applications such as mobile TV, video, and social networking. Considering the

overwhelming demands of mobile data connectivity and its crippling effects on the

Internet bandwidth supported by most mobile operators, the operators are now struggling

to hold their ground by striving for sanity between demands and revenues. The unlimited

(consume all you want) data plans have been very taxing for mobile operators and cannot

be sustained; hence more operators are now inclined towards tiered price plans, imposing

data caps on heavy users and implementing sophisticated traffic management to control

individual users and applications. Many mobile operators are aggressively acquiring new

spectrum and rolling out long-term evolution (LTE) networks in an attempt to increase

capacity and improve the underlying transport economics. Other carriers are searching for

new sources of revenue, such as advertising and data insight, or trying to manage capital

and operational expenses by outsourcing network build and operations, or other new

business models. (Taylor, 2011)

For the last few years, there has been a lot of hype surrounding the Long Term Evolution

(LTE) technology in the mobile operators for increase in capacity after the transformation

from circuit switched to packet switched networks. According to Wireless Intelligence's

2012 report, 500 million LTE connections are predicted globally through over 200 live

LTE networks across more than 70 countries. However, LTE alone cannot solve all the

capacity restraints faced by mobile operators.(Wireless Intelligence's 2012 report)

Alongside the bandwidth requirements of new age Internet applications, customer

demands of mobile data and expectations out of their mobile data experience continue to

rise. End users increasingly expect a highly available, reliable, and fast mobile data

network providing an experience at par with the wireline broadband Internet access.

56

The concept of improving Quality of Service was originally introduced in the

manufacturing as well as service industries mostly as a financially viable investment with

the expectation of resulting returns (Zahorik & Keiningham, 1995). The benefits of

quality improvements come in two forms. One effect is the improved ability of the firm

to attract new customers due to word of mouth, as well as the firm's ability to advertise

the quality of its offerings. The second desired effect is customer retention, which also

increases the word of mouth and reinforces the loop governed by word of mouth. This

observation is also depicted in the aggregated systems dynamics model presented in

Chapter 5 of this thesis.

However, the term Quality of Service is not confined to its traditional business language

(Zahorik & Keiningham, 1995) meaning in this thesis. The term Quality of Service is

taken from language of telecommunications and the Internet, which actually implies the

quality of delivery of data packets (measured through delay, noise, packet loss and

similar parameters) backed by the performance of the network and application layer to

deliver traffic across the Internet to the end user. This traffic is delivered with a pre-

defined level of service that corresponds to certain criteria to counter factors impairing

the quality of Internet traffic (including but not limited to voice, video, and data), such as

packet delay and packet loss etc.

6.2 Quality of Service and Quality of User Experience

There has been a long history of concerns associated with traffic engineering and how

Internet routing is being managed. Its been over a decade since Border Gateway Protocol

(BGP) and Multiprotocol Label Switching (MPLS) were introduced to find efficient and

customized ways of Internet routing. Quality of Service for the Internet is not a new

subject either. It has long been described by the phenomena such as resource allocation

and performance optimization. Traffic engineering mechanisms such as differentiated

services, and integrated services employing RSVP (the resource reservation protocol),

57

are some of the widely known technologies to improve network performance through

resource allocation and performance assurance.

Quality of User Experience (QoE) is another term frequently used in the

telecommunications industry and has been mostly used to evaluate the experience of a

consumer. QoE includes QoS as well as the wider scope of the customer experience

while using an application. QoE is not limited to QoS but also includes social attributes

such as customer's satisfaction level or perception of value. A term called User

Experience Control is used in this thesis as well as in the aggregated systems dynamics

model presented earlier in this thesis. However, this term is used in the context of a

service provider's attempt at shaping the user's experience to ensure a strong relationship

with the customer ensuring customer stickiness, and hence customer loyalty.

Quality of Service for mobile data is specifically described as the availability, reliability,

speed, and throughput available for Internet traffic of any kind on the mobile data

connection. Since this thesis also tackles the application specific monetization of the

Quality of Service of mobile data connectivity, the criteria of availability, reliability,

speed, and throughput are applied to application specific data while considering the

monetization of QoS.

aSma.sRewvs Ewdi Saataduacavau. Eaamo

commnictio (Bok & Hetns 2010)" oni

W-B(-I mdT M hmay M

DdWAMM Abslft4fayinsii

Mbk#&WAWWe VetoIMa (MiAriur Hmwo uum &

Figure 6-1: QoS factors affecting the perceived quality of advanced mobile and IP-based

communication (Brooks & Hestnes, 2010)

58

6.3 Value Networks and Temporariness of Advantage

In his book "Clock Speed", Professor Charles H. Fine talks about the "temporary nature

of advantage", and also states, "The faster an industry evolves - that is, the faster its

clock-speed - the more temporary a company's advantage. The key is to choose the right

advantage - again and again." This statement holds true especially for the "info-

tainment" industry, which has evolved as a result of amalgamation of the multimedia,

information, communications, and electronics (MICE) industries. Analogous to the fiber-

optic communications systems to transmit information and entertainment at the speed of

light, this industry can be considered running at a speed of light (Fine & Kimerling,

1997)

The concept of a value network is particularly relevant to network industries such as

mobile communications and Internet, as well as their amalgamation in the form of

"mobile Internet". Whereas value chains are known to link multiple functions, both intra-

firm and inter-firm level (Porter, 1980, 1985), value networks connect multiple buyers

and sellers at a single node (Normann & Ramirez, 1994; Weiner et al., 1997). Any node

of a value network can be connected to one or many value chains as well as value

networks. The linkage between mobile technology and the Internet itself presents a value

network in addition to being a segment of the traditional telecommunications industry

value chain. The firms owning the interface with the customer (e.g. mobile and fixed line

operators) must often interact with a broad range of buyers, sellers, and providers of

complementary products through multiple platforms (Evans, Hagiu, & Schmalensee,

2006). Amazon.com due to its large alliance program (Afuah & Tucci, 2002), and many

other Internet sites including portals, search engines, auctions, employment sites, online

traders, real-estate sites, and even offer such multiple platforms (Evans et al., 2006) and

can be considered value networks. The idea of layered platforms to implement desired

functionality and the concept of interconnection between providers at different platform

layers (Claffy & Clark, 2013) can further help to understand this notion of value

networks.

59

Figue 62: Te e ergigvlue nwk f h~en Moie inadty( nk 209

ins-I Lcation ServicesGPS

Phone h&Nhny

Mas, Mervices MaltMhsuch Phone 2-D Bar andvidl ladntes afPadket co ercodes

MP4 User Authoriation

aaMb Aotk jut l o mal bPuarisng

is" Dig Wa tellevision stanardSnurt car&s Java-bXked EPCIlfaed

LI ay mreeviceios fo i w mbl hns nyie nwee

GIF: aphieal interface formt R dVrickting Bthr angGPS: glowa positioning sysbunMIDE: Music %nsmumgnt DigiWa IntrfkcEPG: Elecumdni proganting gW&d

Figure 6-2: The emerging value network of the Mobile industry (Funk, 2009)

The above Figure 6-2 shows the expansion and evolution of the mobile industry with not

just the value chain transforming into a value network, but with an inevitable overlap of

the mobile telecommunications with other industries. We have already witnessed the

Mobile Wallet such as Easypaisa*, Mobile Health such as mHealth **, and Mobile phone

home surveillance and security services offered by different mobile operators in the

world.

*Easypaisa Mobile Accounts are actual bank accounts and work just like a normal bank account. This service is available onlyfor Telenor subscribers in certain countries. Using Easypaisa Mobile Account Telenor subscribers can pay bills, transfler money and

use many more services from their own mobile phones. anytimne. anywhere.

**mobile-hecalth (ml--ealth) is the use of mobile computing and communication technologies in health care and public health.

mil ealth has applications Such as facilitating data collection and to encourage health-care consumers to adopt healthy lifestyles or to

self-manage chronic conditions. It can also be used to improve health-care service delivery processes by targeting health-care

providers or communication between these providers and their patients. So, for example, mobile technologies can be used to provide

clinical managemnent support in settings where there are no specialist clinicians, and they can be used to send patients test results and

timely reminders of appointments.

60

As illustrated in Figure 6-2, music, animation, video, and game firms are rapidly evolving

as significant overlapping parts of the mobile phone industry in most countries.

Particularly in Japan, the number and diversity of firms are even larger and include

publishers, retail outlets, restaurants, transportation and travel-related companies,

broadcasters, other general Internet sites, and enterprises that use the Internet to manage

internal operations (Funk,2007). The case study of NTT DoCoMo i-mode discussed in

Chapter 4 of this thesis further provides evidence and insight into this industrial overlap.

The constant zest on the part of mobile operators to enhance their role further from being

the commodity provider / dumb bit pipe carriers to value added service providers for the

users, has been a compelling reason behind launch of these cross industrial value-added

services.

6.4 Value for the mobile Internet user

As discussed in the initial sections of this chapter, the mismatch in demand and capacity,

and the intention to provide user with a better experience, and the desire of network

providers to become more evolved players in the telecommunications value chain are

some of the compelling rationales justifying the monetization of QoS by cellular

operators.

However, the most significant rationale behind monetization of QoS of data is the value

proposition it holds for the mobile user. According to a consumer survey conducted by

Comptel Corporation in 2012, it was discovered that two-thirds of consumers felt

neglected by their mobile operators, and that almost half were likely to chum within the

next two years as a result. The survey also revealed how consumers in United Kingdom,

France, Germany, and the United States would be more loyal and willing to spend more

if operators engaged with them more. The case studies presented in the earlier chapters of

this thesis, back up the concepts of collaborative innovation and user experience control.

Above all, the most significant aspect of these two phenomena is the value perceived by

the end user. Eventually, a technical and/or business venture, whether driven by firm or

collaborative innovation, only makes sense if it provides any prospective value to the end

user.

61

The following Figure 6-3 shows Location of Mobile Data Usage (Percentage of Time

Spent in Activity).

in 6*nmk

Office

Home

Mobil. Intrnt Mobile tig

ane Respondenmsagesd t am*etSource Cisco IBSG Connected Ufe Market Watch. 2011

Figure 6-3: Location based usage of mobile Internet and mobile video (Source Cisco

IBSG, 2011)

It is interesting to note that most of the mobile Internet users, specially the smart phone

users, use stationary wi-fi, or home, office wireless, or wired Internet connectivity for

most percentage of their mobile Internet usage (Schofield, 2013). While being totally

mobile, the speed and availability of their stationary wi-fi connectivity cannot be matched

due to the very nature of mobile connectivity. Even the latest LTE connectivity cannot

match the wired on wireless home and office Internet speeds and throughputs, especially

in areas with congestion and low coverage. With the ever-increasing variety to choose

from, every user has individual preferences of application usage. For example, while

away from home and not connected to home wifi or any public wifi hot spot, some users

may consider location based public transport alert services as being the only critical ones,

while others might prefer their stock exchange application to be delivered at any cost.

62

Despite vast applications of a mobile Internet, and the promise of alacrity and availability

foreseen with the advent of LTE, the cellular data bandwidth is usually always limited by

some constraint of bandwidth as described by the Shannon-Hartley Theorem.

Considering this fact and the temporariness of advantage as mentioned before (Fine, 1998

)implies that the mobile operators are in dire need of utilizing their data capacity in a

more efficient way. This required efficiency can be attained using resource allocation and

performance optimization. Mobile operators are currently using different makeshift as

well as policy-oriented techniques to reduce noise, delay, and jitter in their

communication interfaces. Even in the absence of a widely accepted concept of

monetization of Quality of Service in the context of mobile data connectivity, the

network management, resource allocation, and performance optimization have been

carried out from the User Equipment (UE) to Mobile Base Stations to the core circuit

switched networks.

These in-house and undeclared policies of managing the Quality of Service, which are

developed and implemented by mobile operators independently and in isolation, present a

huge challenge of lack of visibility and uniform standards, as well as regulatory fairness

and compliance. One of the major reasons is that FCC allows the reasonable

discrimination of traffic for network management purposes and it is understandable that

the definition of reasonable usually translates into economically efficient for most mobile

operators. Academics can assist this process by conducting further research on the mobile

Internet QoS management, considering the structure of the industry, the potential revenue

streams, and the processes of standard setting (Funk, 2009).

By making the monetization of Quality of Service transparent, fair, and policy oriented,

users, mobile operators, end users, and regulatory bodies will benefit from enhanced

visibility into the currently grey dimensions of optimum network and resource

management. This transparency will also facilitate the efficient involvement of policy

makers in the process of monetization, both from the economic and regulatory

perspectives. Economists have acknowledged the fact that economic growth of an

63

industry is vastly dependent on the involvement of policy makers to formulate the right

set of policies and rules.

64

Chapter 7

Regulatory Policy

The stakeholders interviewed for this thesis were 50% from the USA, 40% from Europe

and 10% from Middle East and Asia. However, the regulatory constraints addressed in

this thesis are from the US centric point of view for the ease of understanding. Since the

regulation of mobile Internet is still in a nascent stage in the US as well as other parts of

the world, some significant inferences can be drawn and applied to other countries based

on their legislation governing the mobile Internet regulation.

"The "Open Internet" is the Internet as we know it. It 's open because it uses free,

publicly available standards that anyone can access and build to, and it treats all trafic

that flows across the network in roughly the same way. The principle of the Open Internet

is sometimes referred to as "net neutrality." Under this principle, consumers can make

their own choices about what applications and services to use and are free to decide

what lawful content they want to access, create, or share with others. This openness

promotes competition and enables investment and innovation."

Source: www.FCC.gov (2013)

On the FCC website, under the section of The Open Internet rules, it is further

described that one of the significant purposes of the Internet openness is the facilitation it

provides to innovators, application developers and service providers, to introduce novel

technologies and services which can bring drastic improvements to communications as

well as business.

The FCC rules may be general in nature but do specify the fact clearly that the

65

Internet content or applications are not regulated, which implies that the FCC promotes

an environment of open and free competition, cooperation as well as co-opetition in the

Internet world. Another clearly defined objective is that neither the government, nor the

private companies should be given the leverage to deter innovation or its access to the

common Internet user.

7.1 Open Internet Rules

"The FCC has adopted three basic open Internet rules: (www.FCC.gov ,2013)

7.1.a. Transparency

Broadband providers must disclose information regarding their network

management practices, performance, and the commercial terms of their broadband

services.

7.1.b. No Blocking

Fixed broadband providers (such as DSL, cable modem, or fixed wireless

providers) may not block lawful content, applications, services, or non-harmful

devices. Mobile broadband providers may not block lawful websites, or

applications that compete with their voice or video telephony services.

7.1.c. No Unreasonable Discrimination

Fixed broadband providers may not unreasonably discriminate in transmitting

lawful network traffic over a consumer's broadband Internet access service.

Unreasonable discrimination of network traffic could take the form of particular

services or websites appearing slower or degraded in quality."

FCC further recognizes, in the Open Internet Report and Order that "an open,

66

robust, and well-functioning Internet requires that broadband providers have the

flexibility to manage their networks, including but not limited to efforts to block

spain and ensure that heavy users don't crowd out other users. For this reason, the

no blocking and no discrimination rules are subject to reasonable network

management."

The Open Internet R&O by the FCC, has described industry-wide rules to protect

openness for consumers and innovators alike. The Open Internet R&O also recognizes

that mobile broadband presents "special considerations", including the fact that it is

rapidly evolving and has differences from fixed broadband in market structure and

technological constraints. In the advent of 4G technologies, the Commission wants to

ensure the openness for mobile broadband, requiring providers to comply with the

transparency rule, and prohibiting them from blocking websites or from blocking

applications that compete with the provider's voice or video telephony

services. (www.FCC.gov ,2013)

Under the section of industry laws and rules of the FCC website, the Competition section

says,

"The competitive framework for communications services should foster innovation and

offer consumers reliable, meaningful selections in affordable services. The FCC pursues

removing regulatory, economic and operational barriers throughout the

telecommunications sector." (www.FCC.gov ,2013)

The FCC's non-discrimination rule has a broad general objective and it explicitly

aims to ban discrimination, which is "unreasonable", and it also provides a cushion in the

terminology of "exception" for network management purposes. These rules give the

cellular network providers a lot of room and flexibility to formulate the so-called

reasonable discrimination based on the legitimate premise of being justified under

network management requirements.

67

In the USCA Case #11-1355 Document #1415568 Filed: 01/16/2013 Page 31 of 122

where Verizon and MetroPCS (hereafter "Verizon") demanded that the Open Internet

Rules be vacated, the FCC's responses were as follows:

"Broadband providers "shall not unreasonably discriminate in transmitting lawful

network traffic" to their customers. Order 68 (JA 40) (emphasis added); 47 C.F.R. § 8.7.

Network practices are reasonable if they are "tailored to achieving a legitimate network

management purpose," Order 82 (JA 48), such as "ensuring network security and

integrity," contending with "traffic that is unwanted by end users" (by implementing

parental controls), or "reducing or mitigating the effects of congestion on the network."

Ibid. While describing the mobile broadband rules the comments from FCC were: "The

Commission applied even lighter rules to mobile broadband service (e.g., via cellular

networks). Mobile broadband is less mature and more rapidly evolving than fixed

service; consumers have more choices for mobile broadband; and providers face

"operational constraints" that fixed broadband networks do not. Order 94-95 (JA 52-

53). The Commission applied to mobile providers the same transparency rule that applies

to fixed service providers. Order 98 (JA 55). The Commission prohibited mobile

Internet access providers from blocking customer access to lawful websites or

applications that compete with the service providers' own voice or video telephony

services. The Commission declined, however, to apply to mobile service the rule

forbidding unreasonable discrimination, deciding instead to rely on the anti-blocking rule

while continuing to "monitor the development of the mobile broadband marketplace."

Order 104 (JA 58).

FCC's responses from this case clearly show that FCC has given the mobile operators

substantial leverage to cater for the genuine technical constraints of network

management. To ensure that the innovation and development in the mobile broadband

industry is not impaired by regulatory restrictions at this stage, FCC has refrained from

applying stringent rules and has allowed the exceptional flexibility of reasonable traffic

discrimination for efficient network management.

68

Future relationship between Internet service providers and content providers has

been a very controversial aspect of the net neutrality debate (Kramer & Wiewiorra,

2012). Advocates of network neutrality argue that abiding by the stringent network

neutrality rule prohibiting any kind of traffic discrimination is the only way to ensure a

fair competition between content providers. (Lessing, 2001) (Van Schewick, 2006) (Wu

T. , 2003) (Sydell,2006) (Economides & Himmelberg, 1995). The advocates of network

neutrality further believe that it will also compel network operators to enhance their

network capacity to reduce congestion challenges presented by the best-effort internet

trafficking scheme. (Lessing, 2001)

On the contrary, opponents of network neutrality regime argue that QoS based

Internet traffic discrimination will encourage more content variety as well as broadband

investment (Yoo,2005) Regulatory authorities should use academic and industrial

research on the mobile Internet QoS management, considering the structure of the

industry, the potential revenue streams, and the processes of standard setting to formulate

relevant regulatory frameworks. (Funk, 2009). Lahiri, Dewan & Freimer (2013) compare

the two pricing schemes, namely 1) service pricing which involves pricing different

services differently, and 2) traffic pricing which involves pricing the traffic (i.e.,bytes)

transmitted and argue against the general notion that discriminatory prices across services

increase profit for the operator at the cost of consumer benefit. This thesis concurs that

service based/ application specific pricing is beneficial to the consumer as long as it is

available for all applications and content types and the user, rather than the network

operator or the content provider has the choice to discriminate between the applications'

varying QoS levels.

By making the monetization of Quality of Service transparent, fair and policy

oriented, users, mobile operators and the regulatory bodies will benefit from enhanced

visibility into the currently grey world of optimum network and resource management.

FCC acknowledges the challenges and requirements of efficient network management to

meet the mobile data demands and that the mobile operators are bound to manage their

traffic by optimizing resource utilization.

69

This thesis argues that a wide variety of traffic discrimination mechanisms are

already in practice, such as the consumer packages with data caps based on price, the

unavoidable peak hour congestions and performance compromises due to bandwidth

sharing amongst users, resulting in unfair committed interface rate (CIR) due to the best

effort delivery framework. Once the monetization of Quality of Service is identified as a

legitimate practice within the confines of regulatory policy frameworks, it will make the

traffic management and resource optimization transparent. This transparency will also

facilitate the efficient involvement of policy makers in the process of monetization both

from the economic and regulatory perspectives. Economists have acknowledged the fact

that economic growth of an industry is vastly dependent on the involvement of policy

makers to formulate the right set of policies and rules. The isolated network management

policies of managing the Quality of Service, which are currently developed and

implemented by mobile operators independently, represent lack of visibility, uniform

standards, regulatory fairness and hence present huge challenges to end user interests.

One of the major reasons behind this state of affairs is that even through the Net

Neutrality principles apparently forbid discrimination within the Internet traffic, FCC

allows the reasonable discrimination of traffic for network management purposes and it

is understandable that the definition of reasonable usually translates into economically

efficient for most mobile operators.

Barbara Van Schewick (2012) clearly articulates the same question in her paper as how to

best separate socially beneficial and socially harmful discrimination in a way that

(1) Realizes the goals of network neutrality regulation, but

(2) Does not constrain the evolution of the network more than necessary

(3) Makes sense from a policy perspective and is technically sound. (Schewick, 2012)

She further proposes the rule to ban application specific discrimination and to

allow application agnostic discrimination. However, this thesis argues that the

70

application-agnostic discrimination based on user subscription and usage-time has

already been practiced in the form of time-bound bandwidth availability due to

congestion in peak hours, disparity in application structure and consumption combined

with the data usage caps on mobile Internet users. This application agnostic

discrimination, combined with the factors just discussed, leads to an unaccounted form of

application-specific discrimination. The application-specific discrimination over the

Internet implicitly happens or is orchestrated because of the significant differences in the

nature, platform, overhead and development structure of the different applications. The

existence and reality of these differences in the speed and availability of these variegated

applications implies that even application-agnostic but network location-specific, or time-

specific or user-specific discriminations results in "unreasonable" discrimination as a

result of the so-called "reasonable network management requirements"

Even though Barbara Van Schewick (2012) is in favor of only application-

agnostic Quality of Service (QoS) Discrimination, she argues that forms of QoS should

be allowed if they meet the following conditions:

(1) The different classes of service are made available equally to all applications and

classes of applications;

(2) The user is able to choose whether and when to use which class of service

(3) The network provider charges only its own Internet service customers for the use of

the different classes of service. (This restriction would not constrain interconnection

agreements in any way. Thus, payments among interconnecting networks would remain

possible.) Schewick (2012).

An application-specific monetization of Quality of Service can be designed and regulated

based on these three pre-requisites described above where instead of the service provider,

the user should be in charge of the choice of applications/content to be connected as top

priority and the applications/content delivered via the best effort. The consideration,

71

however, remains that the service provider should ensure such a transparent and fair

provision of choice to the consumer using some of the existing and forthcoming

technological capabilities discussed in the next chapter i.e. Chapter 8 of this thesis.

7.2 Regulatory Frameworks for Monetization of QoS

The three case studies presented in Chapters 2,3 and 4 of this thesis present diverse

examples of collaborative innovation and user experience control leading to different

regulatory concerns. Focusing on the specific question of Monetization of QoS of mobile

data, it is imperative to consider at least two different regulatory frameworks as argued

by Barbara Van Schewick (2012), who exposes the deep disconnect between those who,

like most network neutrality proponents and the current FCC, base calls for network

neutrality regulation on a broad theoretical framework that considers a broad range of

economic and non-economic harms", and those who evaluate calls for network neutrality

regulation based on an antitrustframework. Van Schewick (2012) argues that these two

frameworks really lead to very different results regarding which forms of differential

treatment proponents of each framework are concerned about. In the context of the broad

theoretical framework that the FCC explicitly adopts in its order while catering to broad

general rules of Net Neutrality to be adapted by the telecommunications industry, the

FCC's rules make perfect sense. At the same time, network neutrality proponents who are

not aware of the implications of the different frameworks don't necessarily realize that

antitrust-based approaches or approaches that use terms that have well-defined meanings

in antitrust law may reach correct results in the context of an antitrust framework, but do

not capture many instances of discrimination that network neutrality proponents are

concerned about. Thus, adopting such rules would make it impossible to successfully

bring complaints against discriminatory conduct that violates the values that network

neutrality rules are designed to protect. Following two sections view the question of

monetization of QoS of mobile data from these two different lenses of 1) Theoretical/

Open Market and 2) Antitrust regulatory frameworks.

72

7.2.1 Theoretical/Open Market Regulatory Framework

Extending the concept proposed by Barbara Schewick (2012), and framing it to the

application specific QoS discrimination, the first kind of regulatory framework needed

would be theoretical or open Market framework. Such framework would cater to general

transparency and accountability principles applicable to the reasonable data traffic

discrimination allowed for network management. This regulatory framework should

reinforce the key architectural principles on which the Internet was based without stifling

the original architecture of the Internet itself. This kind of framework should serve to

balance the public interest in network neutrality, the benefits that public can reap through

transparent forms of application-specific traffic discrimination, with the legitimate

interests of network providers. It should serve to prevent network providers from

interfering with user choice or distorting competition among applications or classes of

applications, while providing them broad flexibility to differentiate and price their

Internet service offerings and manage their network in efficient and profitable ways. This

framework should enable mobile network providers to offer user-controlled Quality of

Service and provide equal opportunity to all application developers/content providers to

collaborate for the better performance of the network and better quality of the

applications.

As discussed in Chapter 6 while describing the rationale behind monetization of Quality

of Service of Internet data, it is imperative to note that the numerous interviews with

stakeholders from the mobile service providers across US and Europe endorse the fact

that there is a unanimous agreement on the dire need of this allowance in order for mobile

networks to work efficiently and cater for the users' mobile data requirements. It is

advisable for regulatory bodies to understand this technical and business need as most of

the backbone mobile infrastructure providers such as Ericsson and Alcatel are already

advertising the intelligent and efficient management of data traffic to the mobile

operators around the world. With the advent of Long Term Evolution (LTE) technology,

and driven by the technical and business needs, most of the mobile operators across US

73

and Europe have already developed in-house capabilities to discriminate traffic based on

application specific segregation. However, these operators refrain to discuss the current

and future execution plans due to the expected regulatory concerns. In order to come up

with the realistic regulatory policies that address the current technological and business

needs of these mobile operators, the regulatory authorities should consider the theoretical

aspects of the regulatory framework as well as the antitrust aspects of the regulatory

framework

7.2.2 Antitrust Regulatory Framework

As depicted in the case study of Zero.facebook.com's alliance with mobile

operators, which involved business terms that have not been made public, the anti-trust

aspect of regulation for the protection of the consumer's interests and healthy competition

in the market seems very relevant. A similar scenario is observed in Google-Voice and

Sprint case study where collaboration between two apparently competing players can

pose a threat to open competition and impede possibility of further collaboration in the

value chain. Such collaborative initiatives or strategic business alliances can be a source

of concern for the regulatory authorities. Following are some of the recommendations for

the policy makers to identify anti-trust concerns from the inception of a strategic alliance

and also define parameters to apply rules and regulations which can foster business and

technology collaborations and preserve healthy competition in the market.

1. The contractual terms of collaborative ventures should not be fostering

long-term exclusivity in order to eliminate anti-trust concerns. In most

cases, the value chain player with higher stakes desperate to retain their

market share (such as mobile operator) is coerced to form exclusivity

agreements with a player from another part of the value chain (such as

over the top player) to carry a particular application and exclude others.

This is against the FCC's no blocking rule. Such agreement is detrimental

to user interests and the Net Neutrality principles and should be forbidden.

In case of a bit pipe provider, the carrier is not allowed to block any open

74

internet traffic and it should be ensured that all over the top players are

provided equal opportunity to make use of the network operator's internet

transportation resources.

2. It is imperative to ensure that the technical specifications of certain

collaborative integrations should be standardized and announced

publically so that the platform is open and accessible for competitors to

devise similar collaborative integrations to ensure user's openness of

choice. This regulatory assurance will also eliminate the perception of

entrapment as discussed in the Chapter 5 of this thesis. The closeness of

platform and perception of entrapment proves to be detrimental to growth

of the collaborative venture and can prove to be detrimental for one or

both of the players as learned from the Google-Voice and Sprint as well as

NTT DoCoMo i-mode's case studies.

3. Any form of application specific discrimination of Quality of Service

should only be allowed if the user has been provided the interface, the

ability to choose and control enabled by technology to select any

application at any given time to be elevated above the best effort service

level. Any QoS monetization mechanism where such a choice is not

facilitated for the user should be considered a violation of Net Neutrality

principles. The strategy to give the user the choice and control is coherent

with the lessons learnt in the three case studies presented in this thesis.

The regulatory framework's objective is to encourage consumer protection

and progress in a business by creating an environment governed by rules

of healthy competition and innovation. The lessons learnt from the three

case studies consist of promoting collaborative innovation and user

experience control only to the extent that they increase the value for the

user and not beyond the threshold where the perception of intrusiveness

results in reduction of value for end user.

75

Chapter 8

Technology Aspects of Monetization of QoS

"The most profound technologies are those that disappear. They weave themselves into

the fabric of everyday life until they are indistinguishable from it."

-M. Weiser, The Computer for the 21st Century, 1991, p.4 .

This chapter gives a technical perspective of what Internet QoS and mobile

Internet QoS mean and the technology currently available and being orchestrated to

monetize mobile Internet QoS. This chapter is a brief overview of diverse technical

methods available and being developed to manage mobile Internet QoS and may not be

considered comprehensive and definitive since new technology is being rapidly

developed to facilitate this idea.

As discussed in the previous chapter while tackling the regulatory constraints, the

Internet's openness is no doubt its most precious attribute, which also enables its

scalability and extensibility. The following figure illustrates the remarkable openness of

the internet, which allows the freedom to employ any network technology, any type of

middleware software and any type of application and service across the same network.

76

Figure 8-1: National Research Council, Realizing the Information Future, 1994

The National Research Council described this concept as an "open data network

architecture", as long as the users, application developers and service providers all rely

on the "bearer service" of the Internet Protocol- which allows it to transport any load. As

long as the openness of the Internet platform is maintained, the innovations and

77

advancement in and between the application and transport layer should only serve and

not deter the primary objective of Internet evolution.

The Internet can be compared to an urban infrastructure of roads, highways,

tunnels and bridges, alleys and walkways, while the packets travelling over the Internet

can be compared to commuters. The commuters on the Internet use different modes of

transport just like commuters who opt to walk, bike, use sports cars, high efficiency

electric vehicles, SUV's, trucks or trailers and choose their destination. The size and

complexity of the Internet demands that the packets are not discriminated based on the

type of information they are carrying. However, application developers and network

engineers would argue that the Internet traffic transportation is far from fair. The

discrimination may not be packet based, but it definitely is dependent on the network

capacity, congestion and throughput, which combined with the overhead requirements

and specifications of any application, plays a great part in the over all efficiency of the

packet delivery. The following table shows an overview of certain applications and their

QoS requirements.

Application Reliability Delay Jitter Bandwidth

E-Mail High Low Low LowFile Transfer High Low Low Medium

Protocol (FTP)Web access High Medium Low MediumRemote log-in High Medium Medium LowStreaming audio Low Low High MediumStreaming video Low Low High HighVoIP Medium High High LowVideoconferencing Low High High HighMobile TV High High High High

Table 8-1: Applications and corresponding QoS requirements. (Source: Nokia SiemensNetwork 2008: Quality of Service requirements in tomorrow's connected world)

8.1 Internet QoS Today

The Internet traditionally relies on the TCP protocol in the hosts to detect

congestion in the network and reduce the transmission rates accordingly. TCP uses a

78

window-based scheme for congestion-control and the window size corresponds to the

amount of data being transferred between the sender and the receiver. In case of the

detection of a packet loss by the TCP source, it slows down the transmission rate by

slicing the window size to half and then increasing it if and when more bandwidth

becomes available. TCP-based resource allocation requires all applications to use same

congestion control method. However, modification in the TCP stack can enable stealing

into more than a fair share of bandwidth. Also, the best effort service where all packets

are treated equally, is simply not adequate for many new delay sensitive applications.

(Wang, 2001)

8.2 Resource Assurance & Performance Optimization

Over the last decade, the Internet community came up with solutions namely

Integrated Services and Differentiated Services as frameworks to support QoS over the

Internet. While Integrated Services provided resource assurance through resource

reservation for individual application flows, the Differentiated Services amalgamated

edge policing, provision and traffic prioritization. (Wang, 2001)

Resource assurance is the first and performance optimization is the second most

significant building block for supporting QoS over the Internet as it ensures the optimum

and cost-effective resource utilization. Multiprotocol Label Switching (MPLS) provides

the explicit routing capability to gain the necessary control over traffic paths to maximize

resource utilization in the network by setting up virtual circuits in over the IP networks.

Traffic engineering is employed for advanced route selection techniques, also referred to

as constraint-based routing. (Wang, 2001)

8.3 Mobile Internet QoS

8.3.1 Technology Evolution of Mobile Standards

The 3rd Generation Partnership Project (3GPP) unites six telecommunications standard

development organizations (ARIB, ATIS, CCSA, ETSI, TTA, TTC), known as

"Organizational Partners" and provides their members with a stable environment to

79

produce the highly successful Reports and Specifications that define 3GPP technologies.

The 3GPP technologies from these groups are constantly evolving through Generations

of commercial cellular / mobile systems (see Table 8-2 below). Since the completion of

the first Long Term Evolution (LTE) and the Evolved Packet Core specifications, 3GPP

has become the focal point for mobile systems beyond 3G.

The Generations of 3GPP Systems

Generation GSM & 3GPP Major Mfllestones

Ans*me technokogy, from the 1980s .awds.

Vwous technoogIes were dqjyd, Nation y or Region*, intuding:16 #4T (Nordt Mobie Telkphone), AMPS (Advanced Mobie Phone Syste), TACS (Total

Access Communications System), ArNetz to &Net, Radiocom 2000, R"?4 (fadic TulefonoMobile integrato), JTACS (J.pm Total Access Communiions System) and TZ-80n

First digital systems, deployed in the 1990s introducing voice, SMS and data services.The Primary 2G technologies are:

2G GSM/GPRS, CDMAOne, PDC, IDEN, IS-I 36 or D-AMPS.

GSM/GPRS accounts for over 80% of all 2G subscribers (Source:wikipedia&)

Alowed a global vision for the evoknion of, G netwoks, with technologies that ae3G enhancing theNT-2000 faey of systems.

Primay technologies am EDGE (Enhanced OM for GSM), COMA2000 1X/EVDO, UMTS-HSPA+

LTE and LTE-Advanced ham crossed the "generational boundary" offering the nextgeneration(s) of capabilities. With their capacity for high speed data, significant spectralefficiencies and adoption of advanced radio techniques, their emergence is becoming the

beyond 3G basis for all future mobile systems.

It should be noted that LTE-Advanced has qualified as an IiJ-R NT-Advanced radioInterface

Table 8-2: The Generations of 3GPP Systems (Source: www.3gpp.org)

Although these Generations have become an adequate descriptor for the type of network

under discussion, real progress on 3GPP standards is measured by the milestones

achieved in particular Releases as depicted in the table below. New features are

'functionality frozen' and are ready for implementation when a Release is completed.

3GPP works on a number of Releases in parallel, starting future work well in advance of

the completion of the current Release. Although this adds some complexity to the work

of the groups, such a way of working ensures that progress is continuous & stable.

80

Table 8-3 below shows a brief description of various 3GPP Releases prior to and

including the latest technology named LTE (Long Term Evolution).

Table 8-3: 3GPP Release-99 to Release-10

Release 99 - Specified the first UMTS 3G networks, incorporating CDMA air interface.-

March 2000

Release 5 - Introduced mainly IMS, HSDPA (allowing broadband services on the

Downlink), and other minor enhancements.- June 2002

Release 6 - Added HSUPA (enables broadband uploads and services), MBMS, and

enhancements to IMS -Dec 2004

Release 7 - Introduced, among other features, enhancements to High-Speed Packet

Access (HSPA+), QoS, and improvements to real-time applications like VoIP. -Sept

2005

Release 8 - Introduced E-UTRA (also called LTE, based on OFDMA), All-IP Network

(also called SAE), MIMO based network- March 2008

81

Release 9 - SAE enhancements, WiMAX and LTE/UMTS Interoperability- Dec 2008

Release 10 - LTE Advanced fulfilling IMT Advanced 4G requirements.- March 2012

Release 11- E-UTRAN (Evolved Universal Terrestrial Access Network)- Stage 1 freeze

Sept 2011

Release 12- LTE with high spectral efficiency, high peak data rates, short round trip time

and frequency flexibility- Stage 1 freeze March 2012

8.3.2. Long Term Evolution (LTE)

Long Term Evolution (LTE) is the latest mobile standard in which voice in addition to

data to be transferred through packet switched networks. This is a huge contrast to the

previous cellular systems, such as 3G, where voice traffic was transferred through circuit-

switched networks. LTE (Long Term Evolution) or the E-UTRAN (Evolved Universal

Terrestrial Access Network) is the access part of the Evolved Packet System (EPS). The

main requirements for the new access network are high spectral efficiency, high peak

data rates, short round trip time and frequency flexibility.( Nohrborg,2013)

Following are some of the key drivers behind the development and launch of LTE:

a. User demand for higher data rates and quality of service

b. Continued demand for cost reduction (CAPEX and OPEX)

c. Low complexity by avoiding unnecessary fragmentation of technologies

for paired and unpaired band operation

82

Cor

Access,~M

GSM lowS UMTS. T

GEPAN UTRAN ERMAN

Figure 8-2: Evolution from GSM to LTE (Source: www.3gpp.org)

The GSM Mobile Standard was based on circuit switched technology, built to carry real

time services; GSM could support the possibility of data services only through a circuit

switched modem connection, with very low data rates. GPRS was the first leap towards

IP based packet switched technology, with TDMA (Time Division Multiple Access)

employed as air interface and access mechanism, which is similar to GSM.

With the objective of achieving higher data rates and data volume, 3GPP developed

Universal Mobile Telecommunications System (UMTS), which is a Third

generation (3G) mobile cellular system for networks based on the GSM standard.

UMTS was developed with a new access network, based on CDMA (Code Division

Multiple Access). The access network in UMTS operates like a circuit switched

connection for real time services and a packet switched connection for data services (in

black in fig. 8-2 above). In UMTS, an IP address is allocated to the UE (User Equipment)

when a data service is established and released when the service is released. So the initial

data connectivity services still rely upon the circuit switched core.

The Evolved Packet System (EPS) is purely IP based. Both real time and data services

are carried over IP. As soon as the UE /mobile device is switched on, an IP address is

allocated and that IP address is released when that device is switched off. LTE is based

on OFDMA (Orthogonal Frequency Division Multiple Access) to be able to reach even

83

higher data rates and data volumes. High order modulation (up to 64QAM), large

bandwidth (up to 20 MHz) and MIMO (Multiple-Input/Multiple-Output) transmission in

the downlink (up to 4x4) is also a part of the solution. The maximum theoretical data rate

is 170 Mbps in uplink and with MIMO the rate can be as high as 300 Mbps in the

downlink.

The Evolved Packet Core (EPC) is compatible with non-3GPP access technologies such

as WiMAX and WiFi which can be further categorized as Trusted and Non-Trusted

access networks. This categorization is based on the business relation/agreement between

these networks and not on the technical grounds.

The LTE access network is simply a network of base stations, Evolved NodeB (eNB),

generating a flat architecture. There is no centralized intelligent controller, and the eNBs

are normally inter-connected by the X2-interface and towards the core network by the

Si-interface. The reason for distributing the intelligence amongst the base-stations in

LTE is to speed up the connection set-up and reduce the time required for a handover.

For an end-user the connection set-up time for a real time data session is in many cases

crucial, especially in on-line gaming. The time for a handover is essential for real-time

services where end-users tend to end calls if the handover takes too long.

Beyond 3G wireless systems will support real-time (inelastic bandwidth demands), and

non real-time (elastic bandwidth demands) traffic over a common packet-switched radio

channel using QoS mechanisms to differentiate the traffic. A key challenge in accurately

modeling user and system performance is to be able to simultaneously account for the

impacts of mixed traffic types, QoS mechanism and varying radio channel. Over the last

few years, researchers and engineers, have come up with numerous strategies, including

traffic routing design and programming for the purpose of extending and combining a

number of techniques into a single approximate analytic methodology for providing fast

and accurate performance analysis of multiple inelastic and elastic traffic classes, subject

to pre-emptive priority for the inelastic traffic.

84

Beyond 3G wireless broadband systems namely LTE and LTE-Advanced support a

diverse range of real-time and non real-time services using only packet switching

techniques, without the conventional dedicated channels of 3G and 2G systems. They

will therefore use Quality of Service (QoS) schemes based on preemptive priority (3GPP

TS 23.107 V6.4.0 & Alasti, Neekzad, Hui & Vanithamy, 2010) to ensure that the real-

time or rate-"inelastic" traffic (e.g. voice over IP and video) has sufficiently high

bandwidth and low delay for acceptable user quality. At the same time the systems

should maximize the aggregate throughput (i.e. system efficiency) while also delivering

practically usable performance for the non real-time or rate-"elastic" traffic types (e.g.

web surfing, e-mail or file downloads). Beyond 3G wireless systems use "adaptive

modulation and coding" (AMC) to maximize the data rate of each "call", depending on

the "signal to interference and noise ratio"(SINR) experienced. Thus these systems

exhibit variable capacity on the air interface. For the inelastic traffic this means that the

radio resources (e.g. resource blocks in LTE) allocated to each connected session will

vary depending on the SINR. On the other hand, the resources allocated to elastic calls

are subject to variations in traffic patterns of the higher priority inelastic traffic, as well as

the variable capacity radio channel. Hence the connected session could be (i) a real-time

inelastic user session whose duration is driven only by holding time e.g. a voice call, and

(ii) a non real- time elastic user session whose duration is driven by the combination of

session volume in bytes and the available capacity, e.g. data transfer. A key issue in the

design of these next generation wireless systems is how to evaluate the user and system

performance for realistic user traffic comprising both inelastic and elastic calls at a range

of loads. There have been a lot of work done in this field in terms of modeling and testing

policies and mechanisms to distinguish elastic and inelastic data /type of traffic and then

apply automatic rules to manage traffic optimization to ensure resource segregation and

performance optimization based on alleviating the inelastic and elastic traffic from

affecting the other. The same techniques can be applied to application-specific traffic

optimizations based on similar schemes conceptualized from user-requested requirements

whereas the access interface enables the orchestration of this choice to be user-controlled

and hence fair and transparent. A few of such mechanisms are described in the sections

below.

85

8.3.3 Cognitive Radio Admission Control Scheme

In order to provide QoS requirements in high speed future communication networks, such

as LTE, the operator has to provide a Radio Admission Control algorithm which will

guarantee the QoS of different service types (e.g. voice, data, video, ftp) while

maximizing radio resource utilization. One such scheme is the Cognitive Radio

Admission Control Scheme based on a Multilayer Feed-forward Neural Network.

According to this scheme, the eNodeB performs Radio Admission Control using a

cognitive engine that learns from the past experience how the admission of a new session

would affect the QoS of all sessions in the future. (Bojovic, Baldo & Dini, 2011)

8.3.4 Network and Device aware Quality of Service

No matter what the service is, users will always expect powerful, sound and stable

functions. For multimedia videos, stability is of the greatest importance. Users expect to

watch videos smoothly and at a certain level of quality, no matter what changes occur in

the network environment. However, the existing video platforms often provide

inconsistent playback, resulting from the fluctuation of network on-line quality,

especially with mobile devices, which have limited bandwidth and terminal unit hardware

resources. As the number of network users is rapidly increasing, bandwidth insufficiency

will occur and then network multimedia services will be affected significantly. In contrast

to general services that have a high acceptance rate for packet loss, multimedia packets

emphasize the correctness, sequence order and real-time nature of packets. When a

multimedia video service standard is applied, the service quality declines greatly while

trying to meet the demands of video transmission. Users often view live videos that

freeze, have intermittent sound, or even fail to operate. Therefore, how to execute smooth

playback with limited bandwidth and the different hardware specifications of mobile

streaming is an interesting challenge. (Wang, Lai, Chao & Nan, 2011)

8.3.5 Interactive Mobile Multimedia Services

86

H.264/SVC is an extended coding and decoding architecture based on H.264/AVC. The

benefit of H.264/SVC is that it can adjust the image quality dynamically, according to the

bandwidth of the receiving end. The draft was proposed in April of 2004 and was decided

on in November of 2007. SVC puts forward a brand-new layered architecture. This

hierarchical structure can realize the scalability of temporal, spatial and quality

dimensions. The spirit of SVC is that the receiving end is sure to receive image packets

of the lowest quality for decoding. The image layer with the lowest quality is called the

base layer. The base layer of SVC is completely compatible with H.264/AVC, and when

there is enough bandwidth to receive image packets with higher quality, the decoder will

carry out reference decoding according to the received packets, which is to say, high

quality image packets cannot resolve images independently; the image packet of the base

layer must be consulted for decoding. In terms of the scalability of the three dimensions,

SVC uses the hierarchical B-picture method to realize temporal scalability, down/up

sampling filters and inter-layer prediction to realize spatial scalability, and Signal-to-

Noise Ratio (SNR) scalability and a Metal Gear Solid (MGS) Codec to realize quality

scalability. The scalability of hierarchies in the video can be determined during the course

of coding. (Wang, Lai, Chao & Nan, 2011)

In addition, interactive mobile multimedia services communicate and coordinate the

mobile device with the server-side to select the multimedia file applicable to the device

environment (bandwidth, resolution and arithmetic capability) (Psannis, Ishibashi,

Hadjinicolaou, 2007), (Saranya & Vijayalakshmi, 2011) so as to realize an optimal

multimedia streaming service. This previous study proposed an interactive mobile

multimedia service over cloud computing, which was introduced in (Lai & Vasilakos,

2010), (Chang & Huang, 2012) In the previous service, the mobile device side exchanges

information with the cloud environment, so as to determine an optimum multimedia

video. Scholars have done numerous researches toward conventional platform, which is

the Content Data Network (CDN) to store different movie formats in a multimedia server,

to choose the right video stream according to the current network situation or the

hardware calculation capabilities. To solve this problem, many researches have attempted

dynamic encoding to transfer media content, but still cannot offer the best video quality.

87

This is due to the time consuming fact that traditional encoding requires re-coding of the

entire multimedia content. (Lai, Wang, Chao & Nan, 2011)

8.4 Bearers & QoS - EPS QoS Parameters

8.4.1 Default & Dedicated Bearer

One EPS bearer is established when the UE connects to a Public Data Network (PDN)

and it remains established throughout PDN lifetime. This is default bearer. Any additional

EPS bearer for the same PDN is referred to as a dedicated bearer.

8.4.2 GBR Vs. Non-GBR

An EPS bearer is referred to a GBR if it offers a Guaranteed Bit Rate, else it would be a

Non-GBR bearer. Each bearer has an associated QoS parameter. Default EPS bearer is

always non-GBR, where as dedicated can be either.

Following table shows the Quality Control Indicator levels in LTE where GBR stands for

Guaranteed Bit Rate and Non-GBR stands for Non-Guaranteed Bit Rate.

Since Bearers are established in LTE as discussed above, the priority levels with

corresponding packet delay and packet error loss rates are depicted in the table. QoS

can be enforced and controlled per EPS bearer or a set of aggregated EPS Bearers.

8.4.3 QCI - QoS Class Identifier

QCI is used as a reference to access specific parameters that control how bearer level

packets are forwarded.

88

Bemrrs & QoS: EPS QoS: QCI Service Characteristics

1 GR 2 100 10 cnversitimai

2 4 1:5 t05CW~ttmaVo(LvJftai3 5 300 tor Non-CcavemfimalVio

4 3 50 la, Real Tine Gming

S 1 100 tal IMSSfti

6 Non-GBR 7 100 10 Vie.,Vdeo Uve Strearaing

7 6 300 10 vudeo (fldfned sk""lins)

9 1300 IV TCP (Emai, ChsC p)

9300 10- Fie Mritg

Table 8-4: Bearers & QoS: QCI Service Characteristics

8.4.4 PCRF (Policy & Charging Rules Functions)

- Policy & Charging Rules Functions (PCRF) encompasses business and

technological rules that govern which network services a subscriber can access at

what bandwidth level, when and for how long.

- It aggregates information in real time to support creation of rules and

automatically making intelligent policy decisions for active subscribers.

8.5 Network Application Plug-Ins (APIs)

Open operator APIs is the terminology that describes exposure of the operator's

internal assets to third parties across well-defined open interfaces. . Such APIs can serve

the vastly diverse purposes ranging from enhanced network QoS delivery for video

content through subscriber's data management assets for profiling and advertising. The

objective is to provide the content and application developers with an opportunity to offer

a better service than would be possible in the regular over the top environment and with

89

no access to the operator's internal assets. The following figure shows some developer

programs that make use of, and encourage development, to open APIs.

CO~ PROV~M Ce0ilf

" t Wdesiined devei pq WeM wir 30 AM% pubtobd;0"'"9* includA" 0rengV Apio" SWo

Swiamcm OpwnAfis sm#eeeornfrvieclmngmn ppK*Achas dik 1* a

Telecom Next Open Inchudes Telecom API A Device SOK progrns;la Innovaonprmiy a te ib wbi Med bae d users

TkelIa WM MopAfpwuste4 AETin prtFIls

VO" BeWavrn Betn is a sme1-sxae deveopment program fht inckude a fmw opeVoaoeVodbboe 3M0 AOW;Vodefmn 360 offers APIs baWe oa A. hnked to OWePI

Table 8-5: Examples of APIs Operator/Developer Programs (Source: Brown, 2012)

In contrast to the profitability witnessed in a closely integrated collaborative initiative

taken by NTT DoCoMo while completely tying the content providers to its proprietary

platform, here the revenue sharing terms are not as immediately promising. While there

are not many API-oriented services generating substantial revenues today; there is a

certainly a long-term strategic promise of value associated with these initiatives. At the

very least, such services help operators form closer ties with the third party content

providers, and help the operators attain a better position in the evolution of the

telecommunications value chain where their status could be elevated from being the silent

by-standers who (without such Network API's) have no linkage with the content

developers and hence no standing ground to progress towards a fairer share in the future

telecommunications value chain.

It is imperative to note that this thesis presents a brief overview of the various

current and future technical approaches to manage mobile Internet QoS. Additionally,

new technological solutions are being rapidly developed to facilitate management and

monetization of QoS.

90

Chapter 9

Future Work

9.1 Collaborative Innovation & User ExperienceControl

"Innovation- any new idea-by definition will not be accepted at first. It takes repeated

attempts, endless demonstrations, monotonous rehearsals before innovation can be

accepted and internalized by an organization. This requires courageous patience."

- Warren Bennis

This notion about reiterative improvements holds even stronger in the case of

collaborative innovation, as the two or more firms collaborating have to break internal

inertia as well as the combined inertia of the collaborative venture, iteratively, till the

collaboration failure rate is reduced to a minimum (refer to Figure 5-3 & Figure 5-4,

Chapter 5). The collaborative innovation ventures and strategic business alliances as

depicted from the zero.facebook.com, NTT DoCoMo and Google Voice-Sprint case

studies will be replicated more in the near future marking the transformation of the

telecommunications ecosystem and the evolution of the traditional value chain into a vlue

network as explained in Chapter 1 and Chapter 5 of this thesis.

One such example was very recently reported by the Wall Street Journal on June

26, 2013 from New Delhi, India, that Google Inc. and Bharti Airtel Ltd. have announced

the joint venture launch of the "Free zone," or free Internet, for first time mobile Internet

users in India, which happens to be the world's second largest telecom market, (Krishna,

91

2012). This venture is also targeted for a price-sensitive customer base by offering a low

cost value service feature for the consumers. Voice connectivity has become a

commodity in emerging and competitive mobile markets of Africa and South Asia, where

mobile Internet services are generating the most revenue as the tariffs for voice calls are

amongst the lowest in the world due to the fierce competition and price wars between

incumbent and entrant mobile operators amidst their efforts to claim the available market

share. The free Internet zones will enable Airtel mobile customers access to mobile web

search and will feature phone-friendly versions of Google's email and Google+ social

networking services. This is a very recent example of the subtle application of Quality of

Service discrimination, on the same lines as was witnessed in the case of

zero.facebook.com.

"Free Zone aims to put the web in the hands of more people and empower first time

Internet users with several useful services of the Internet," the two companies said in a

statement. "The first page of a website linked from search results is provided at no data

cost." (Krishna, 2012).

Aligned with the common theme of the three case studies evaluated in this thesis, this

recent example of the Free Zone collaborative venture not only represents an example of

the mobile operators getting into the content side of the Internet, but also shows the

inevitable application-specific discrimination in a subtle way. It is noteworthy that,

similar to the zero.facebook.com case, the viewers will be getting a flavor of Google

email, Google + for social networking and some searching and surfing for free, in certain

Free zones offered by a telecom operator. This is also a salient example of shaping

consumer experience. The developed and saturated markets such as US and Europe

should foresee the monetization of QoS of mobile data as an inevitable transition or at

least an incubatory step which will lead to the evolution of the telecommunications value

chain where the mobile operators will be involved in the user's Over the Top experience

of content and applications in a more meaningful way. With the right regulatory policies

in place, this involvement can be made fair, transparent and beneficial for the user as well

as for the service provider.

92

From the three cases studies, the themes of collaborative innovation and user experience

control were extrapolated and explored in the aggregated model of this thesis. However,

it is important to understand that some times the collaborative venture has little to do with

innovation in technology and more to do with bundling of two or more services to shape

the user experience. This was observed in the three case studies and is analogous to the

example of popcorn and movies. The two products are developed independently.

However, their businesses are complimenting each other. When more people go to the

movies and buy expensive popcorn at the movie theatre, the popcorn sales increase.

However, it is still essential to make good movies and market them well for more people

to be willing to go to the movie theatre. Similarly, the popcorn quality has to be

maintained at a certain level for people to keep associating a good movie experience with

popcorn consumption.

9.2 LTE & Small Cell Mobile Technology

While the Release 8 of LTE as described in Chapter 8, has enabled the packet-based

transportation for voice and data alike, the future development in mobile technology

governed by 3GPP is being orchestrated based on the data traffic needs and management.

The enhancements for small cells have now been specifically prioritized as a key

component of Release 12 of the 3GPP standard. The mobile operators are considering

numerous ways to increase capacity to meet the data explosion; with more spectrum,

interworking with other access technologies, or through the use of new antenna

technologies. All have a role to play, but Matthew Baker identified as 'vital' the role of

small cells (Baker, 2013)

Addressing the Small Cells World Summit, in London on June 2 5th 2013, Baker said

"Small cells using 3GPP radio access technologies will enhance capacity and per-user

throughput, as well as reducing costs and uniquely offering tight cooperation with the

macro coverage layer. Further optimization is coming, especially aimed at dense

93

deployments with large numbers of small cells in hot-zones." The maximum potential

benefits of the Small Cells enhancement in the mobile technology can only be reaped

with the monetization of Quality of Service of mobile data.

9.3 QoE Management Framework

In Chapter 6, section 6.2 of this thesis, some differentiation was established

between Quality of Service (QoS) management and Quality of Experience (QoE)

management. The next evolutionary step from the monetization of QoS for mobile data is

the management of Quality of Experience. Most of the mobile operators are already

conducting quality and connectivity surveys and data mining analysis to measure Quality

of experience for voice services mainly. This concept should be further applied to Voice,

Video and Data, once the mobile operator has the required insight and QoS visibility into

the user's data connectivity experience.

To correctly measure the "user experience" of service quality, appropriate QoE metrics

should be defined. The measurement metrics should include factors that indicate quality

levels. QoS metrics indicate the availability, reliability and quality, each for reflections

among service accessibility and latency on network traffics. The quality of service can be

measured with the video resolution, application response time as well as user feedback

and rating measures currently utilized by the content evaluation processes in place. The

QoE metrics would essentially also include objective QoS parameters such as bandwidth,

delay, loss or jitter, and human factors such as emotions and experience (ITU-T, 2008).

For specific service areas such as video streaming, services may include application level

QoE metrics such as frame rate, resolution (ITU-T, 2008), re-buffering events

(Huysegems, Vleeschauwer & Schepper, 2012), interactive delay (Yuan, Zimu &

Baochun, 2011) and interruption (Gheibi, Medrad & Ozdaglar, 2011)

Each service is different from each other, and different characteristics may require

different metrics for QoE measurement. For example, a service, which requires data to be

downloaded, such as FTP, may regard the bandwidth as the most important measurement

metric. A video service such as IPTV may need complex quality metrics for file transfer,

94

and may include jitter as a key QoE metric. For the above reason, it is difficult to choose

an optimum set of metrics that is suitable for all services. We may first consider all

possible QoE metrics before QoE measurement, and may remove unimportant factors

after evaluating their influence on a specific service. While broadening the QoS

measurement and monetization by including the social elements of the experience and

encompassing the evaluation and management of Quality of Experience, QoE

management frameworks can be developed to amalgamate the additional experience

elements and correlated the Quality of Service with the other parameters affecting the

user experience.

9.4 The Aggregated Model

The aggregated model presented in the Figure 5-1 and Figure 5-2 (Chapter 5) is

considered in this thesis for one value chain player at a time. For future work and further

extrapolations, sensitivity analyses and testing policy levers such as regulatory techniques

and service feature life cycles, it is noteworthy, that the aggregated model can be used

simultaneously for two or more different players of the telecommunications value chain.

The model presented here can be further utilized by considering it in the form of super

imposed layers when being employed to analyze the parameters for more than one value

chain player simultaneously. For example, the collaborative failure rate of two value

chain players can be analyzed for the third player's measurement of barriers to new

entrants and the expected consumer surplus share can be evaluated as a sum of expected

revenue for two service providers operating in collaboration or coopetition.

9.5 The Regulatory Policy Recommendations

The regulatory concerns and the debate around application-agnostic and application-

specific Quality of Service management for mobile data has been discussed in the 7th

95

Chapter of this thesis in detail. However, to conclude, it is strongly recommended for

regulatory authorities and policy makers in the US in particular, and Europe and the rest

of the world in general to acknowledge the monetization of QoS of mobile internet as an

existing technical capability which is constantly being enhanced with upcoming mobile

advancements such as Long Term Evolution (LTE) and others discussed in Chapter 8,

and an impending business reality which has been considered seriously by mobile

operators and content providers for quite some time now. In the interest of innovation,

generating better value for the end user and efficient network resource management, the

regulatory authorities should play an effective part by acknowledging this phenomenon

and developing policies to ensure better visibility, transparency and accountability within

this realm.

96

APPENDIX A

Questionnaire for interviews with stakeholders across the telecommunications

industry

o Voice becoming commodity?

o 'All you can eat data' bundles sustainable?

o Premium charges by service based traffic segregation a possibility?

o What are the new generation services currently being focused for leveraging existing investments

in the network infrastructure?

o Cloud based services a major focus for CSPs? If so what are some of major corporate, consumer

services being offered now? In near future?

o Any plans to move towards carrier services?

o If mobility not a core service, are there plans to add mobility? Own network or MVNO?

o If fixed line not a core service, are there any plans to add fixed? Cable / Copper / Fiber? Own

network or partnership with an existing fixed player?

o Is convergence (fixed / mobile) a key for survival?

o What is the main hurdle in offering new services to customers?

o What is the main revenue driver in the Internet / Voice / Video bundles?

o What percentage of revenue can be attributed to Voice? Data? TV/Video? What's the trend in 3-

5 years?

o How's interconnect revenue distribution changing?

o Is consolidation the way forward to achieve economies of scale? Have reached the point in

telecom spiral where consolidation is inevitable for cost control and revenue sustainability?

97

o Key investment focus on network, user experience, service offering flexibility?

o Investment in 4G (LTE...) essential for survival for wireless operators?

o Investment in spectrum is on the up?

o Acquisition part of the next 3-5 year strategy? If so for customer base? economies of scale?

Spectrum?

o Main hook for new customers?

o Main hook for existing customers to stay?

o What is the net additions now? Target for next 3-5 years?

o Current Churn?

o Main reasons for Chum?

o Main customer segment Churning?

o Key strategy for controlling Chum? Price? Service? Technology?

o How's the competition is changing with advent of outside players?

o What is the key strategy to counter that? Offer similar services? Cloud based services? Service

based traffic charging?

o What's the emerging role for device manufacturers?

Are the subscriptions to TV bundles going down with online (mostly free) TV services?

Cable / Satellite TV is a major revenue pipe? What's future of Cable?

Cable survival based on TV or Internet?

98

0

0

0

Demand for ubiquitous connectivity?

Core, transmission and access network capacity?

4G spectrum acquisition and network upgrade have a business case? Or survival case?

o On demand on the rise?

o IPTV becoming mainstream? What's the impact?

Broadband still being offered as commodity fixed cost bundles? What's the future?

Consumers willing to pay for premium speed? Access?

What is the current revenue share for content based services? On the rise?

CSP's share in the content revenue?

o Google / Skype encroaching on the CSP revenue? Future trend?

o CSP becoming dumb pipe providers?

What is the key regulation shift foreseeable?

CSP should be allowed to charge premium for certain traffic type (VoIP traffic)?

o In terms of service offerings?

o Regulatory framework?

o Customer trend?

o Investment strategy?

99

0

0

0

0

0

0

0

0

0

APPENDIX B

Causal Trees for Some Parameters in the Aggregated Model

Profitability of Service Provider

capital per service feature

effect of capital on innovation ratereference capital per service feature innovation rate

profitability Lookup

Normal innovation rate

INITIAL TIME Time

change adaptability by userBarriers to New Entrants

(customer stickiness)customer stickiness

Existing customers

Competitive AdvantageC t Perception of Value for Money

Platform Enrichment

100

Perception of Value for MoneyA doption rate

Word of Mouth

(Time)

(Total Customer base)

User Experience Control >change adaptability by user

Perception of Intrusiveness Total Customer base

Time

Barriers to New Entrantscustomer stickiness

(Existing customers)

(Perception of Value for Money)

Existing customers

INITIAL TIME Time

Adoption rate

(change adaptability by user)Total Customer base

customer stickiness

Existing customers change adaptability by user

User Experience Control Loss rateUser Experience Control

user experience control rate

(User Experience Control)Perception of Intrusiveness

consumer profiling

Competitive AdvantagePerception of Value for Money

Platform Enrichment Adoption rate

customer stickiness Word of Mouth

101

Time

Barriers to New Entrantscustomer stickiness Word of Mouth

Existing customers

Perception of Value for Money

User Experience Control Loss rate

User Experience Controluser experience control rate Perception of Intrusiveness

customer stickiness consumer profiling

User Experience ControlPerception of Intrusiveness

consumer profilingUser Experience Control Loss rate

Competitive AdvantagePerception of Value for Money

Platform Enrichment

Perception of IntrusivenessUser Experience Control Loss rate

Perception of Value for Money

Billing Interface Ownership User Experience Control

consumer profiling user experience control rate

Platform Enrichment

102

Time

(Total Customer base)

Billing Interface Ownership

collaborative innovation rateProfitability of Service Provider

Customer maintenance OPEX

Firm Innovation Rate

Total Customer base >RevenueARPU

Collaborative Innovation

Firm Innovation Competitive Advantage

competitors avg innovation level Perception of Value for Money

(Collaborative Innovation)

Platform Enrichment

(Firm Innovation)

103

INITIAL TIME Time

Adoption rate

change adaptability by userTotal Customer base

customer stickiness

Existing customers

(Time) Cost of service provider

Firm Innovation Billing Interface Ownership

(innovation rate) collaborative innovation rate

Customer maintenance OPEX

Collaborative InnovationFirm Innovation Rate

innovation rate

104

APPENDIX C

Equations for basic run stimulation of the aggregated model

1. Adoption rate=Perception of Value for Money+Word of Mouth

Units: cust/year

2. ARPU=

"Consumer spending (on QoS etc)"+Platform Enrichment

Units: $/year

3. Barriers to New Entrants= INTEG (customer stickiness-change adaptability by

user,0)

Units: dmnl [0,10]

4. Billing Interface Ownership=

Firm Innovation/Time

Units: dmnl/year

5. capital per service feature=

RAMP(max capital/(FINAL TIME-INITIAL TIME), INITIAL TIME,

FINAL TIME)

Units: $/service feature

6. change adaptability by user=

(Perception of Intrusiveness-User Experience Control)* Total Customer base

*Time

Units: cust/year

7. Collaboration Failure rate=

innovation obsoletion rate + integration faults

105

Units: dmnl/Month [0,10]

8. Collaborative Innovation= INTEG (collaborative innovation rate-Collaboration

Failure rate,0)

Units: service features [0,100]

9. Collaborative innovation rate= 100 * innovation rate

Units: service features/year

10. Competitive Advantage=

Collaborative Innovation Firm Innovation-competitors average innovation level

Units: service features [0,100]

11. Competitors average innovation level= constant value of 50 inferred from market

data on service features of mobile service features including package plans etc

Units: service features

12. consumer profiling=0.8*customer stickiness (various fractions of customer

stickiness were tried)

Units: dmnl [0,10]

13. Consumer spending (on QoS etc)= Perception of Value for Money

Units: $/Month [0,10]

14. Cost of service provider=

(Billing Interface Ownership * Total Customer base)/Time + Firm Innovation

Rate

+ 0.5* collaborative innovation rate+(Total Customer base

*Customer maintenance OPEX)

Units: $/year

106

15. Customer maintenance OPEX= MAX[OPEX, BILLING INT

OWNERSHIP+OPEX] is the operational cost of maintaining one customer per

year

Units: $/year

16. Customer stickiness=

(Barriers to New Entrants+Perception of Value for Money)*Existing customers

*Time

Units: cust/Month [0,10]

17. Effect of capital on innovation rate=

Profitability Lookup (capital per service feature/reference capital per

service feature)*0. 1* Profitability of Service Provider

Units: dmnl

18. Existing customers=

le+06

Units: customers

19. FINAL TIME = 2053

Units: year

20. The final time for the simulation.

Firm Innovation= INTEG (Firm Innovation Rate-firm innovation failure rate,l)

Units: service features

21. Firm innovation failure rate= constant values tested ranging 0 to 10

Units: service features/year [0,10]

22. Firm Innovation Rate= ACTIVE INITIAL (

RAMP(2,0,(3 *innovation rate + 0.3 *Collaborative Innovation)),

107

10)

Units: service features/year [0,50,1]

23. INITIAL TIME = 2013

Units: year

The initial time for the simulation.

24. Innovation obsoletion rate=RAMP (1,0,0.1)*Time

Units: service features/year

25. Innovation rate=

Normal innovation rate*effect of capital on innovation rate

Units: service features/$/year

26. Integration faults=

I to 30

Units: service features/year

27. Max capital= fraction of profitability

Units: $/service feature

28. Normal innovation rate

Units: service features/year

29. Perception of Intrusiveness=

0.5*consumer profiling + 0.5*User Experience Control

Units: dmnl

30. Perception of Value for Money=

Competitive Advantage*Platform Enrichment

Units: $/service feature [0,100]

108

31. Platform Enrichment=

Collaborative Innovation + Firm Innovation

Units: service features [0,10]

32. Profitability Lookup(

[(0,0)-(30,100)],(0.794297,4.7619),(6.17108,19.523 8),(12.4644,30),(16.2525

,3 5.2381),(19.8574,47.1429),(24.1955,62.381),(27.2505,62.3 81),(29.6334,63.809

5))

Units: dmnl

33. Profitability of Service Provider-- INTEG (

Revenue-Cost of service provider,0)

Units: $ [0,10]

34. Reference capital per service feature= various constant values tested

Units: $/service feature

35. Revenue=

ARPU*Total Customer base

Units: $/year

36. Total Customer base= INTEG (

Adoption rate + customer stickiness*Total Customer base-change

adaptability by user*Total Customer base, Existing customers)

Units: customers

37. User Experience Control= INTEG (user experience control rate-User Experience

Control Loss rate, 1)

Units: dmnl

109

38. User Experience Control Loss rate=

Perception of Intrusiveness-Perception of Value for Money

Units: dmnl/Month

39. User experience control rate=

Billing Interface Ownership+ consumer profiling +Platform Enrichment

Units: dmnl/Month

40. Word of Mouth=

0.5 * customer stickiness

Units: cust/Month

110

References

1ST-1999-10050. (2013). BRAIN (Broadband Radio Access for IP-based Networks). INT

J COMPUT COMMUN, 8 (1), 127-135.

3GPP. (2006). Quality of Service (QoS) Concept and Architecture. 3GPP.

3GPP RANI Chairman, B. M. (2013, June 25). From Macro to Small Cells:

Enhancements for Small Cells in 3GPP. (L. Small Cells World Summit, Interviewer)

Afuah, A., & Tucci, C. (2002). Internet Business Models and Strategies: Text and Cases.

New York: McGraw-Hill / Irwin.

Alasti, M., Neekzad, B., Hui, J., & Vanithamy, R. (2010). Quality of Service in WiMAX

and LTE Networks. IEEE Communications Magazine, 48 (5).

Andersson, B., Malinen, J., Forsberg, D., Kari, H., Hautio, J., Mustonen, K., et al. (n.d.).

Dynamics Mobile IP. Retrieved from http://dynamics.sourceforge.net

Appling, G. (2011). Telecom Value Chain Disruption. London, United Kingdom: The

Financial Times Limited.

Beller, A., Jamhour, E., & Pellenz, M. (2004). Defining Reusable Business-Level QoS

Policies for DiffServ. Proceedings of Distributed Systems Operations and Management

Workshop, (pp. 40-51).

Benmammar, B., & Krief, F. (2007). QoS Management for Mobile Users. 229.

Blackburn, B. (2011, March 11). Japan Earthquake & Tsunami: Social Media Spreads

News, Raises ReliefFunds. Retrieved June 27, 2013, from ABC News/Technology:

http://abcnews.go.com/Technology/japan-earthquake-tsunami-drive-social-media-

dialogue/story?id=13117677

111

Blake, S., Black, D., & Carlson, M. (1998). An Architecture for Differentiated Services.

IETF RFC 2475.

Bless, R., Zitterbart, M., Hillebrand, J., & Prehofer, C. (2003). Quality-of-Service

Signaling in Wireless IP-based Mobile Networks. VTC2003-Fall. Orlando: VTC.

Bojovic, B., Baldo, N., & Dini, P. (n.d.). A Cognitive Scheme for Radio Admission

Control in LTE Systems. Centre Tecnologic de Telecomunicacions de Catalunya (CTTC)

Bougard, B. (2005). Article title not available. IEEE .

Boyle, J. (1996). Shammans, Software, and Spleens: Law and Construction of the

Information Society (4 ed.). Cambridge, Massachusetts, U.S.A.: Harvard University

Press.

Braun, T., & Stattenberger, G. (2003). Providing Differentiated Services to Mobile IP

Users. The 26th Annual IEEE Conference on Local Computer Networks (LCN 2001).

Tampa: IEEE.

Brooks, P., & Hestnes, B. (2010). User Measures of Quality of Experience: Why Being

Objective and Quantitative is Important. IEEE Network, 24 (2), 8-13.

Brown, G. (2010). Monetization Strategiesfor Mobile Operators. Heavy Reading on

behalf of Cisco.

Cabral, L. (1990). On the Adoption of Innovations with 'Network' Externalities.

Mathematical Social Sciences, 19, 299-308.

Carr, N. (2010). The Shallows: What the Internet is Doing to Our Brains. New York,

New York, U.S.A.: W.W. Norton.

112

Chan, K., Seligson, J., Durham, D., Gai, S., McCloghrie, K., Herzog, S., et al. (2001).

COPS Usage for Policy Provisioning (COPS-PR). IETF RFC 3084.

Chang, S., Lai, C., & Huang, Y. (In Press). Dynamic Adjustable Multimedia Streaming

Service Architecture over Cloud Computing. Computer Communications.

Chin-Feng, L., & Honggang, W. (2011, January). A Network and Device Aware QoS

Approach for Cloud-Based Mobile Streaming. IEEE Transactions on Multimedia.

Cisco. (2011). Cisco IBSG Connected Life Market Watch. Cisco.

Cisco. (2011). Cisco Virtual Networking Index.

Claffy, K., & Clark, D. (2013). Finding Theory for a Dynamic Internet. Unpublished

work- please contact authors kc@caida.org and ddc@csail.mit.edu.

Clements, M., & Ohashi, H. (2005). Indirect network Effects and the Product Cycle:

Video Games in the U.S., 1994-2002. Journal ofIndustrial Economics, 53 (4), 515-542.

Curwen, P., & Whalley, J. (2010). Mobile Telecommunications in a High-Speed World:

Industry Structure, Strategic Behavior and Socio-Economic Impact.

de Geus, A. (1988). Planning as Learning. Harvard Business Review, 66, 70-74.

Drager, D. (2011). Lifehacker.com. Retrieved from http://lifehacker.com/5781404/the-

five-best-voip-or-voice-chat-apps

Dranove, D., & Gandal, N. (2003). The DVD vs. DIVX Standard War: Empirical

Evidence of Network Effects and Preannouncement Effects. Journal ofEconomics &

Management Strategy, 12 (3), 363-386.

Dyrud, M. A. (2011). Social Networking and Business Communication Pedagogy:

Plugging into the Facebook Generation. Oregon Institute of Technology, Business

Communication Quarterly, 74 (4), 475-478.

113

Economides, N., & Himmelberg, C. (1995). Critical Mass and Network Evolution in

Telecommunications Industry. In G. Brock (Ed.), Selected Papersfrom the 1994

Telecommunications Policy Research Conference.

Eisenberg, J. K., & Liddle, D. E. Wireless Technology Prospects and Policy Options.

Resultsfrom a National Academies Study. National Research Council (NRC). NRC.

European Regulators Group. (2007). Opinion on Functional Separation.

Evans, D., Hagiu, A., & Schmalensee, R. (2006). Invisible Engines. Cambridge: MIT

Press.

Eyre, R. (2008). New Content Can Trigger Mobile Advertising Bonanza. Media Week

(1158), 21.

Farrell, J., & Saloner, G. (1985). Standardization, Compatibility, and Innovation. The

RAND Journal of Economics, 16 (1), 70-83.

FCC. (2010). Preserving the Open Internet: Broadband Industry Practices, 25. FCC.

Federal Communications Commission. (n.d.). Retrieved from

http://www.fcc.gov/guides/open-Internet

Federal Communications Commission. (2010). Retrieved December 12, 2012, from

http://hraunfoss.fcc.gov/edocspublic/attachmatch/FCC- 10-201 A1_Rcd.pdf

Feng, Y., Zimu, L., & Baochun, L. (2011). Gesture Flow: QoE-Aware Srtreaming of

Multi-Touch Gestures in Interactive Multimedia Applications. IEEE Journal on Selected

Areas in Communications, 30 (7), 1281-1294.

Fine, C. (1999). Clockspeed: Winning Industry Control in the Age of Temporary

Advantage. Reading, Massachusetts, U.S.A.: Perseus Books.

114

Fine, C., & Kimerling, L. (1997). Biography of a Killer Technology: Optoelectronics

Drives Industrial Growth with the Speed of Light. Optoelectronics Industry Development

Association.

Frieden, R. (n.d.). Invoking & Avoiding the First Amendment: How Internet Service

Providers Leverage Their Status as Both Content Creators and Neutral Conduits.

Funk, J. (2011). Journal of Engineering and Technology Management, 28, 232-248.

Funk, J. L. (2009). The Emerging Value Network in the Mobile Phone Industry: The

Case of Japan and its Implications for the Rest of the World. Telecommunications Policy

,33, 4-18.

Funk, J. (2007). Solving the Startup Problem in Western Mobile Internet Markets.

Telecommunications Policy , 14 (1), 14-30.

Gandal, N. (2000). The Dynamics of Technological Adoption in Hardware / Software

Systems: The Case of Compact Disc Players. The RAND Journal of Economics, 31 (1),

43-61.

Garcia-Murillo, M., & Gabel, D. (2003). International Broadband Deployment: The

Impact of Unbundling. 31st Telecommunications Policy Research Conference. Arlington,

VA.

Globalization of Mobile and Wireless Communications Signals and Communication

Technology. (2011). pp. 119-128.

Grindley, P. (1995). Standards Strategy and Policy: Cases and Stories. Oxford: Oxford

University Press.

Huysegems, R., De Vleeschauwer, B., & De Schepper, K. e. (June 4-5, 2012). Session

Reconstruction for HTTP Adaptive Streaming - Laying the Foundation for Network

115

based QOe mONITORING. Proceedings of the 20th International Woekshop on Quality

ofService, (pp. 1-9). Coimbra, Portugal.

International Telecommunication Union. (2010).

ITU-T. (2008). ITU Telecommunication Standardization Sector (ITU-T): Objective

Perceptual Multimedia Video Quality Measurement in the Presence of a Full Reference,

ITU-T Recommendation. ITU-T.

ITU-T. (2008). ITU Telecommunication Standardization Sector. Quality of Experience

Requirements for IPTV Services, ITU-T Recommendations. ITU-T.

Ivanovich, M., & Fitzpatrick, P. An Accurate Performance Approximation for Beyond 3G

Wireless Broadband System with QoS. IEEE.

Jamhour, E., Fonseca, M., Beller, A., & Pereira, T. (n.d.). Imaculada Concei9ao 1155.

Retrieved from PPGIA: http://www.ppgia.pucpr.br/docentes.html#redes

Jelassi, S., Rubino, G., Melvin, H., & Pujolle, G. (2012). Quality of Experience of VoIP

Service: A Survey of Assessment Approaches and Open Issues. Communications

Surveys & Tutorials, 14 (2), 491-513.

Jordan, S. (2010). Do Wireless Networks Merit Different Net Neutrality Than Wired

Networks?

Jorguseski, L. (n.d.). Vision on Radio Resource Management (RRM) and Quality of

Service (QoS) for Wireless Communication Systems in Year 2020.

Kapor, M. (n.d.). The Software Design Mnifesto. Retrieved from

http://www.kei.com/homepages/mkapor/SoftwareDesignManifesto.html

Kim, T. (2006). Priority Based Delay Mitigation. IEEE Communications Letters.

116

Kincaid, J. (2011, June 23). Facebook Now has 750 Million Users. Retrieved June 26,

2013, from TechCrunch: http://techcrunch.com/2011/06/23/facebook-750-million-users/

Kirkpatrick, D. (2010). The Facebook Effect: The Inside Story of the Company that is

Connecting the World New York: Simon & Schuster.

Koubaa, A. (2010). Guaranteeing Real Time Service. IEEE Transactions on Industrial

Electronics .

Krishna, R. (2013, June 26). Google, Bharti to Offer Free Mobile Internet Services. The

Wall Street Journal.

Lai, C., & Vasilakos, A. (2010). Mobile Multimedia Services over Cloud Computing.

IEEE COMSOC MMTC E-Letter, 5 (6).

Leonard-Barton, D. (1995). Wellsprings of Knowledged. Boston: Harvard Business

School Press.

Lessig, L. (1999). Four Puzzles from Cyberspace. Code and Other Laws of Cyberspace .

Lima, A., Suave, J., & Souza, N. (2012). Capturing the Quality and Business Value of IT

Services Using a Business-Driven Model. Network and Service Management, IEE

Transactions on, 9 (4), 421-432.

Lyons, D. A. (n.d.). Net Neutrality and Nondiscrimination Norms in

Telecommunications: A Historical Perspective.

Meejoung, K. (2010). Priority Based Service Difference. IEEE Transactions on

Vehicular Technology.

Mullins, J. (2007). The New Business Road Test. London: Pearson.

Nair, H., Chintagunta, P., & Dube, J. (2004). Empirical Analysis of Indirect Effects in the

Market for personal Digital Assistants. Quantitative Marketing and Economics, 2, 23-58.

117

National Broadband Network. (2009, April). Regulatory Reform for 21st Century

Broadband Discussion Paper. Australia: Government of Australia.

National Cable & Telecom Association vs. Brand X Internet Servs., 545 U.S. 967 (2005).

Gaiti, D. (Ed.). (2005). Network Control and Engineering for QoS, Security and

Mobility. Fourth IFIP International Conference on Network Control and Engineering for

QoS, Security and Mobility. Lannion, France.

Nohrborg, M. (2013). 3GPP. Retrieved from http://www.3gpp.org/LTE

Normann, R., & Ramirez, R. (1994). Designing Interactive Strategy: From the Value

Chain to the Value Cancellation. New York: Wiley.

Ohashi, H. (2003). The Role of Network Effects in the US VCR Market, 1978-1986.

Journal of Economics & Management Strategy, 12 (4), 447-494.

Oomori, T. (2008, August 15). iPhone Sheds Light on Weakness of Japanese Handset

Makers. Nikkei Electronics.

Pack, S., & Choi, Y. (2001). An End-to-End QoS Provisioning Architecture in Mobile

Network. Proc. International Symposium on Communications and Information

Technologies (ISCIT). Chiangmai, Thailand.

Parandeh-Gheibi, A., Medard, M., & Ozdaglar, A. (2011). Avoiding Interruptions- A

QoE Reliability Function for Streaming Media Applications. IEEE Journal on Selected

Areas in Communications, 29 (5), 1064-1074.

Pareek, D. (2006). Business Intelligence for Telecommunications. Auerbach Publications.

Perkins, C. (1996). IP Mobility Support. IETF RFC 2002.

Pisano, G., & Verganti, R. (2008). What Kind of Collaboration is Right for You?

Harvard Business Review, 78, 86.

118

Porter, M. (1985 ). Competitive Advantage: Creating and Sustaining Superior

Performance. New York: New York Free Press.

Porter, M. (1980). Competitive Strategy: Techniques for Analyzing Industries and

Competitors. New York: New York Free Press.

Posner, R. A. (2010). New Republic, 241 (13), 27-3 1.

Prasad, R., Dixit, S., van Nee, R., & Ojanpera, T. (n.d.). Globalization of Mobile and

Wireless Communications - Today and in 2020. ISBN: 978-94-007-0106-9 (Print).

Psannis, K., & Ishibashi, Y. H. (2007). QoS for Wireless Interactive Multimedia

Streaming. ACM Workshop on QoS and Security for Wirelss and Mobile Networks, (pp.

168-171).

Rayburn, D. (2012). Mobile Application Accelaration. Frost & Sullivan.

Rohlfs, J. (1974). A Theory of Interdependent Demand for a Communications Service.

Bell Journal of Economics and Management Sciences, 5 (1), 16-37.

Rohlfs, J. (2001). Bandwagon Effects in high-Technology Industries. Cambridge,

Massachusetts, U.S.A.: MIT Press.

Rust, R. T., Zahorik, A. J., & Keiningham, T. L. (1995). Return on Quality (ROQ):

Making Service Quality Financially Accountable. Journal ofMarketing, 59 (2), 58.

Rust, R., Zahorik, A., & Keiningham, T. (1995). Return on Quality (ROQ): Making

Service Quality Financially Accountable. Journal of Marketing, 59 (2), 58.

Samak, T., & Al-Shaer, E. (2012). Fuzzy Conflict Analysis for QoS Policy Parameters in

DiffServ Networks. Network and Service Management, IEEE Transactions on, 9 (4),

459-472.

119

Saranya, S., & Vijayalakshmi, M. (2011). Interactive Mobile Live Video Learning

System in Cloud Environment. International Conference on Recent Trends in

Information Technology (ICR TIT), (pp. 673-677).

Schnizlein, J., Strassner, J., Scherling, M., Quinn, B., Herzog, S., Huynh, A., et al.

(2001). Terminology for Policy-Based Management. IETF RFC 3084.

Schofiled, J. (2013, June 12). Most Mobile Data Will Soon be Offloaded to Wi-Fi

Networks, says Juniper Research.

Selby, J. (n.d.). Anyone's Game: A Market-Structure Based Approach to Economic and

Regulatory Implications of an Internet of Things.

Shah, A., Sicker, D. C., & Hatfield, D. N. (n.d.). Thinking About Openness in th

Telecommunications Policy Context.

Shapiro, C., & Varian, H. (1999). Information Rules. Boston: Harvard Business School

Press.

Shin, S. (n.d.). A Novel Method for Service Differentiation in IEEE 802.15.4: Priority

Jamming.

Sicker, D., & Mindel, J. Refinements of a Layered Model for Telecommunications

Policy. Journal on Telecommunications and High Technology Law, 1, 69-94.

Snir, Y., Ramberg, Y., Strassner, J., Cohen, R., & Moore, B. (2003). Policy Quality of

Service (QoS) Information Model. IETF RFC 3644.

Stuckwisch, M. (2008, January). Wikipedia. Retrieved from en.wikipedia.org.

Tam, D. (2013, January 30). CNET News. Retrieved July 19, 2013, from

http://news.cnet.com/8301-1023_3-57566550-93/facebook-by-the-numbers-1.06-billion-

monthly-active-users/

120

Taylor, S. (2011, November). A New Chapter for Mobile? How Wi-Fi Will Change the

Mobile Industry As We Know It. Cisco Internet Business Solutions Group (JBSG) .

The Nielsen Company. (2010, August 2). What Americasn Do Online: Social Media and

Games Dominate Activity. Retrieved June 26, 2013, from Nielsenwire:

http://blog.nielsen.com/nielsenwire/onlinemobile/what-americans-do-online-social-

media-and-games-dominate-activity/

(1999). Third Report and Order and Fourth Further Notice of Proposed Rulemaking, In

the Matter of Implementation of the Local Competition Provisions of the

Telecommunications Act of 1996.

van Schewick, B. (2012, June 11). Network Neutrality and Quality of Service: What a

Non-Discrimination Rule Should Look Like.

Wang, Z. (2001). Internet QoS, Architectures and Mechanisms for Quality of Service.

Waterman, D., & Choi, S. (2010, August 3). Network Neutrality and Vertical Control:

Lessons from Cable TV. Bloomingdale, Indiana.

Waxman, H. A. (2010, September 29). Waxman Statement on Net Neutrality Proposal.

Retrieved January 2012, from http://waxman.house.gov/chairman-waxmans-statement-

net-neutrality-proposal

Weiser, P. J. (2008). The Next Frontier for Network Neutrality. Administrative Law

Review, 60 (2), 273-322.

Weiser, P. J. (2003). Toward a Next Generation Regulatory Strategy. Loyola University

of Chicago Law Journal, 35 (1), 41-85.

Werbach, K. (2003, August 29). SUPERCOMMONS: Towards a Unified Theory of

Wireless Communication.

121

Werner, T. (2012, May 15). Comcast Voices. (Comcast) Retrieved January 2013, from

blog.comcast.com: http://blog.comcast.com/2012/05/the-facts-about-xfinity-tv-and-xbox-

360-comcast-is-not-prioritizing.html

Wernick, C. (2007). Strategic Investment Decisions in Regulated Markets. The

Relationship between Infrastructure Investments and Regulation in European

Broadband Deutscher Universitats-Verlag.

Wiener, M., Nohria, N., Hickman, A., & Smite, H. (1997). Value Networks - The Future

of US Electric Utility Industry. Sloan Management Review, 38, 21-40.

Wu, I., McElvane, R., Dey, A., & Duwadi, K. (2003, September 11). Current

International Issues in Communications Regulation: Refining Regulatory Techniques,

Impact of Wireless' Popularity on Regulation, and the Growing Significance of Cable

Television.

Wu, T. (2006). Why Have a Telecommunications Law? Anti-Discrimination Norms in

Communications. Journal on Telecommunications & High Tech. Law, 15.

Yang, L. T., Chen, T. M., Zhang, Y., & Mao, S. (2008). Broadband Mobile Multimedia:

Techniques and Applications. Auerbach Publications.

Zhang, J., & Hwa-Jong, K. User Feedback Oriented Quality of Experience Management

Framework. Kangwon National University, Department of Computer Engineering.

Gangwondo: Kangwon National University.

122